Listings
All listings: Microsoft Azure
Every listing for the models on page 2 of 13 of Microsoft Azure, including each model's other versions and routes. Back to Microsoft Azure
Sold by Microsoft Azure
44 listings on this page sold by Microsoft Azure.
Select a score for details.
About these columns
- llminfo IQ
- llminfo IQ (capability index). How llminfo IQ is computed
- Coding IQ
- llminfo Coding IQ (coding capability, its own scale). How Coding IQ is computed
- Value
- Value for money. 50 means as expected for its results. Above 50, cheaper per task. How Value for money works
- Yap
- Tokens used per coding task vs. models with the same results. Includes agents that write code. Chattiness on maths doesn't predict it. How the Yap score works
- Epoch Capabilities Index
- Epoch Capabilities Index benchmark details
| Model | llminfo IQ | Coding IQ | Input $/1M | Output $/1M | Max Input Tokens | Value | Yap | Epoch Capabilities Index |
|---|---|---|---|---|---|---|---|---|
Grok 4.7 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 137.7From 249 benchmarks. Plenty of data behind it. | 117 ±7Coding IQ 117.4, 95% interval 111–124. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.00>200k $4.00cache read $0.50 | $6.00>200k $12.00 | 500K | 41.2Poor value (41.2). Costs more per task than its benchmark results predict. Plenty of data behind it. | 55 ±14Typical (55 ±14). Generates about as many tokens as models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate. | 156.4Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Claude Opus 4.6 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 136.9From 466 benchmarks. Plenty of data behind it. | 118 ±5Coding IQ 117.9, 95% interval 113–123. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $5.00cache read $0.50 | $25.00cache write $6.25 | 1M | 47.2Poor value (47.2). Costs more per task than its benchmark results predict. Plenty of data behind it. | 41 ±15Terse (41 ±15). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 156.3Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
136.8From 489 benchmarks. Plenty of data behind it. | 112 ±8Coding IQ 112.0, 95% interval 104–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.75cache read $0.275batch $1.375 | $16.50batch $8.25 | 922K | 60.1Good value (60.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 28 ±8Terse (28 ±8). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 157.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
136.8From 489 benchmarks. Plenty of data behind it. | 112 ±8Coding IQ 112.0, 95% interval 104–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.75cache read $0.275batch $1.375 | $16.50batch $8.25 | 922K | 60.1Good value (60.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 28 ±8Terse (28 ±8). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 157.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
Claude Sonnet 5 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 136.3From 360 benchmarks. Plenty of data behind it. | 115 ±5Coding IQ 115.2, 95% interval 111–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.00cache read $0.20 | $10.00cache write $2.50 | 1M | 37.4Poor value (37.4). Costs more per task than its benchmark results predict. Plenty of data behind it. | 57 ±9Chatty (57 ±9). Generates more tokens than models with similar coding results. Measured at medium effort. Little data behind it, so treat it as a rough estimate. | 156.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5.6 Terra 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 135.7From 329 benchmarks. Plenty of data behind it. | 121 ±5Coding IQ 120.9, 95% interval 116–126. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.20cache read $0.22 | $13.20cache write $2.75 | 922K | 60.4Good value (60.4). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 48 ±10Typical (48 ±10). Generates about as many tokens as models with similar coding results. Measured at max effort. A fair amount of data behind it. | 160.8Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5.6 Terra 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 135.7From 329 benchmarks. Plenty of data behind it. | 121 ±5Coding IQ 120.9, 95% interval 116–126. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.00cache read $0.20 | $12.00cache write $2.50 | 922K | 60.4Good value (60.4). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 48 ±10Typical (48 ±10). Generates about as many tokens as models with similar coding results. Measured at max effort. A fair amount of data behind it. | 160.8Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-6 Luna 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 135.6From 227 benchmarks. Plenty of data behind it. | 117 ±6Coding IQ 116.9, 95% interval 111–123. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.12cache read $0.012 | $0.60cache write $0.15 | 922K | 63.4Good value (63.4). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 51 ±11Typical (51 ±11). Generates about as many tokens as models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate. | 156.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-6 Luna 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 135.6From 227 benchmarks. Plenty of data behind it. | 117 ±6Coding IQ 116.9, 95% interval 111–123. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.10cache read $0.01 | $0.50cache write $0.125 | 922K | 63.4Good value (63.4). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 51 ±11Typical (51 ±11). Generates about as many tokens as models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate. | 156.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5.6 Luna 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 135.2From 344 benchmarks. Plenty of data behind it. | 112 ±10Coding IQ 112.4, 95% interval 102–123. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.22cache read $0.022 | $1.32cache write $0.275 | 922K | 66.2Good value (66.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 52 ±8Typical (52 ±8). Generates about as many tokens as models with similar coding results. Measured at max effort. A fair amount of data behind it. | 156.7Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5.6 Luna 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 135.2From 344 benchmarks. Plenty of data behind it. | 112 ±10Coding IQ 112.4, 95% interval 102–123. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.20cache read $0.02 | $1.20cache write $0.25 | 922K | 66.2Good value (66.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 52 ±8Typical (52 ±8). Generates about as many tokens as models with similar coding results. Measured at max effort. A fair amount of data behind it. | 156.7Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5.2 Pro 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 134.7From 42 benchmarks. Plenty of data behind it. | —Not enough data yet | $21.00batch $10.50 | $168.00batch $84.00 | 272K | 49.6Fair value (49.6). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | 157.4Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5.2 Pro 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 134.7From 42 benchmarks. Plenty of data behind it. | —Not enough data yet | $10.50 | $84.00 | —Not published by the provider | 49.6Fair value (49.6). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | 157.4Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5.2 Pro 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 134.7From 42 benchmarks. Plenty of data behind it. | —Not enough data yet | $21.00batch $10.50 | $168.00batch $84.00 | 272K | 49.6Fair value (49.6). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | 157.4Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Claude Haiku 5.5 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 134.6From 230 benchmarks. Plenty of data behind it. | 124 ±6Coding IQ 123.6, 95% interval 118–129. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.10cache read $0.01 | $0.50cache write $0.125 | 1M | 62.1Good value (62.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 49 ±9Typical (49 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 160.3Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Claude Opus 4.5 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 134.0From 384 benchmarks. Plenty of data behind it. | 113 ±5Coding IQ 112.8, 95% interval 108–118. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $5.00cache read $0.50 | $25.00cache write $6.25 | 200K | 46.0Poor value (46.0). Costs more per task than its benchmark results predict. Plenty of data behind it. | 43 ±10Terse (43 ±10). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 151.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5.1 Codex Max 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. 🇦🇺🇧🇷+17 australia-east, brazil-south +23🇦🇺🇧🇷+17 australia-east, brazil-south, canada-east, us-central, us-east, us-east-2, france-central, germany-west-central, japan-east, korea-central, us-north-central, europe-north, norway-east, poland-central, south-africa-north, us-south-central, south-india, spain-central, sweden-central, switzerland-north, uae-north, united-kingdom-south, europe-west, us-west, us-west-3 | 134.0From 11 benchmarks. A fair amount of data behind it. | 112 ±9Coding IQ 111.7, 95% interval 103–121. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.25cache read $0.125 | $10.00 | 272K | 47.0Poor value (47.0). Costs more per task than its benchmark results predict. A fair amount of data behind it. | 48 ±9Typical (48 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | —Not measured |
133.9From 361 benchmarks. Plenty of data behind it. | 113 ±9Coding IQ 113.3, 95% interval 105–122. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.54cache read $0.15 | $4.84 | 1M | 45.9Poor value (45.9). Costs more per task than its benchmark results predict. Plenty of data behind it. | 58 ±10Chatty (58 ±10). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 153.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
133.9From 361 benchmarks. Plenty of data behind it. | 113 ±9Coding IQ 113.3, 95% interval 105–122. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.31cache read $0.231 | $7.26 | 1M | 45.9Poor value (45.9). Costs more per task than its benchmark results predict. Plenty of data behind it. | 58 ±10Chatty (58 ±10). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 153.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
Claude Sonnet 4.6 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 133.8From 469 benchmarks. Plenty of data behind it. | 108 ±6Coding IQ 108.4, 95% interval 103–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $3.00cache read $0.30 | $15.00cache write $3.75 | 1M | 35.4Poor value (35.4). Costs more per task than its benchmark results predict. Plenty of data behind it. | 71 ±8Chatty (71 ±8). Generates more tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 152.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GLM 5.3 Flash 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 133.6From 266 benchmarks. Plenty of data behind it. | 113 ±8Coding IQ 112.8, 95% interval 105–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.188cache read $0.038 | $0.625 | 1M | 65.3Good value (65.3). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 68 ±11Chatty (68 ±11). Generates more tokens than models with similar coding results. Measured at max effort. A fair amount of data behind it. | 153.8Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5.3 Codex 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. 🇦🇺🇧🇷+17 australia-east, brazil-south +23🇦🇺🇧🇷+17 australia-east, brazil-south, canada-east, us-central, us-east, us-east-2, france-central, germany-west-central, japan-east, korea-central, us-north-central, europe-north, norway-east, poland-central, south-africa-north, us-south-central, south-india, spain-central, sweden-central, switzerland-north, uae-north, united-kingdom-south, europe-west, us-west, us-west-3 | 133.6From 51 benchmarks. Plenty of data behind it. | 114 ±8Coding IQ 113.7, 95% interval 106–122. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.75cache read $0.175 | $14.00 | 272K | 59.7Good value (59.7). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 46 ±9Typical (46 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | —Not measured |
GPT 5.2 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version🇦🇺🇧🇷+16 australia-east, brazil-south +22🇦🇺🇧🇷+16 australia-east, brazil-south, canada-east, us-central, us-east, us-east-2, france-central, germany-west-central, japan-east, korea-central, us-north-central, norway-east, poland-central, south-africa-north, us-south-central, south-india, spain-central, sweden-central, switzerland-north, uae-north, united-kingdom-south, europe-west, us-west, us-west-3 | 133.4From 440 benchmarks. Plenty of data behind it. | 112 ±7Coding IQ 111.5, 95% interval 105–118. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.75cache read $0.175batch $0.875 | $14.00batch $7.00 | 272K | 56.6Good value (56.6). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 49 ±14Typical (49 ±14). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 154.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
133.4From 440 benchmarks. Plenty of data behind it. | 112 ±7Coding IQ 111.5, 95% interval 105–118. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.925cache read $0.1925 | $15.40 | —Not published by the provider | 56.6Good value (56.6). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 49 ±14Typical (49 ±14). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 154.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
133.4From 440 benchmarks. Plenty of data behind it. | 112 ±7Coding IQ 111.5, 95% interval 105–118. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.75cache read $0.175batch $0.875 | $14.00batch $7.00 | 272K | 56.6Good value (56.6). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 49 ±14Typical (49 ±14). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 154.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
133.4From 440 benchmarks. Plenty of data behind it. | 112 ±7Coding IQ 111.5, 95% interval 105–118. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.875 | $7.00 | —Not published by the provider | 56.6Good value (56.6). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 49 ±14Typical (49 ±14). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 154.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
133.4From 440 benchmarks. Plenty of data behind it. | 112 ±7Coding IQ 111.5, 95% interval 105–118. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.875 | $7.00 | —Not published by the provider | 56.6Good value (56.6). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 49 ±14Typical (49 ±14). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 154.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5 Pro 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version🇦🇺🇧🇷+17 australia-east, brazil-south +22🇦🇺🇧🇷+17 australia-east, brazil-south, canada-east, us-east, us-east-2, france-central, germany-west-central, japan-east, korea-central, us-north-central, europe-north, norway-east, poland-central, south-africa-north, us-south-central, south-india, spain-central, sweden-central, switzerland-north, uae-north, united-kingdom-south, europe-west, us-west, us-west-3 | 132.8From 14 benchmarks. Plenty of data behind it. | —Not enough data yet | $15.00batch $7.50 | $120.00batch $60.00 | 272K | 29.6Poor value (29.6). Costs more per task than its benchmark results predict. A fair amount of data behind it. | —Not measured | 151.3Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Kimi K2.6 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. Newer version | 132.6From 305 benchmarks. Plenty of data behind it. | 113 ±7Coding IQ 112.9, 95% interval 106–119. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.045cache read $0.176 | $4.40 | 262K | 50.4Fair value (50.4). Costs per task about what its benchmark results predict. Plenty of data behind it. | 48 ±14Typical (48 ±14). Generates about as many tokens as models with similar coding results. Measured at default effort. A fair amount of data behind it. | 151.6Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
DeepSeek V4 Flash 0731 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. Newer version | 132.2From 149 benchmarks. Plenty of data behind it. | 113 ±8Coding IQ 112.8, 95% interval 105–121. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.44cache read $0.014 | $1.32 | 1M | 64.8Good value (64.8). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 57 ±9Chatty (57 ±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 154.4Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5.2 Codex 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version🇦🇺🇧🇷+17 australia-east, brazil-south +23🇦🇺🇧🇷+17 australia-east, brazil-south, canada-east, us-central, us-east, us-east-2, france-central, germany-west-central, japan-east, korea-central, us-north-central, europe-north, norway-east, poland-central, south-africa-north, us-south-central, south-india, spain-central, sweden-central, switzerland-north, uae-north, united-kingdom-south, europe-west, us-west, us-west-3 | 131.9From 52 benchmarks. Plenty of data behind it. | 110 ±6Coding IQ 109.8, 95% interval 104–116. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.75cache read $0.175 | $14.00 | 272K | 52.6Good value (52.6). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 59 ±11Chatty (59 ±11). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | —Not measured |
GPT 5 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version🇺🇸🇫🇷+6 us-east, us-east-2 +11🇺🇸🇫🇷+6 us-east, us-east-2, france-central, germany-west-central, us-north-central, europe-north, poland-central, us-south-central, spain-central, sweden-central, europe-west, us-west, us-west-3 | 131.7From 584 benchmarks. Plenty of data behind it. | 111 ±5Coding IQ 110.8, 95% interval 106–116. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.375cache read $0.1375batch $0.625 | $11.00batch $5.00 | 272K | 52.5Fair value (52.5). Costs per task about what its benchmark results predict. Plenty of data behind it. | 65 ±8Chatty (65 ±8). Generates more tokens than models with similar coding results. Measured at medium effort. A fair amount of data behind it. | 150Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT 5 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer versiongpt-5-2025-08-07 · 2025-08-07 · 🇦🇺🇧🇷+17 australia-east, brazil-south +22gpt-5-2025-08-07 · 2025-08-07 · 🇦🇺🇧🇷+17 australia-east, brazil-south, canada-east, us-east, us-east-2, france-central, germany-west-central, japan-east, korea-central, us-north-central, europe-north, norway-east, poland-central, south-africa-north, us-south-central, south-india, spain-central, sweden-central, switzerland-north, uae-north, united-kingdom-south, europe-west, us-west, us-west-3 | 131.7From 584 benchmarks. Plenty of data behind it. | 111 ±5Coding IQ 110.8, 95% interval 106–116. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.25cache read $0.125batch $0.6875 | $10.00batch $5.50 | 272K | 52.5Fair value (52.5). Costs per task about what its benchmark results predict. Plenty of data behind it. | 65 ±8Chatty (65 ±8). Generates more tokens than models with similar coding results. Measured at medium effort. A fair amount of data behind it. | 150Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
131.7From 584 benchmarks. Plenty of data behind it. | 111 ±5Coding IQ 110.8, 95% interval 106–116. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.625 | $5.00 | —Not published by the provider | 52.5Fair value (52.5). Costs per task about what its benchmark results predict. Plenty of data behind it. | 65 ±8Chatty (65 ±8). Generates more tokens than models with similar coding results. Measured at medium effort. A fair amount of data behind it. | 150Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
131.7From 584 benchmarks. Plenty of data behind it. | 111 ±5Coding IQ 110.8, 95% interval 106–116. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.625 | $5.00 | —Not published by the provider | 52.5Fair value (52.5). Costs per task about what its benchmark results predict. Plenty of data behind it. | 65 ±8Chatty (65 ±8). Generates more tokens than models with similar coding results. Measured at medium effort. A fair amount of data behind it. | 150Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT 5.1 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version🇦🇺🇧🇷+17 australia-east, brazil-south +23🇦🇺🇧🇷+17 australia-east, brazil-south, canada-east, us-central, us-east, us-east-2, france-central, germany-west-central, japan-east, korea-central, us-north-central, europe-north, norway-east, poland-central, south-africa-north, us-south-central, south-india, spain-central, sweden-central, switzerland-north, uae-north, united-kingdom-south, europe-west, us-west, us-west-3 | 131.0From 352 benchmarks. Plenty of data behind it. | 107 ±8Coding IQ 106.6, 95% interval 99–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.25cache read $0.125batch $0.6875 | $10.00batch $5.50 | 272K | 49.7Fair value (49.7). Costs per task about what its benchmark results predict. Plenty of data behind it. | 50 ±9Typical (50 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 150.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
131.0From 352 benchmarks. Plenty of data behind it. | 107 ±8Coding IQ 106.6, 95% interval 99–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.375cache read $0.1375 | $11.00 | —Not published by the provider | 49.7Fair value (49.7). Costs per task about what its benchmark results predict. Plenty of data behind it. | 50 ±9Typical (50 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 150.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
131.0From 352 benchmarks. Plenty of data behind it. | 107 ±8Coding IQ 106.6, 95% interval 99–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.625 | $5.00 | —Not published by the provider | 49.7Fair value (49.7). Costs per task about what its benchmark results predict. Plenty of data behind it. | 50 ±9Typical (50 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 150.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
131.0From 352 benchmarks. Plenty of data behind it. | 107 ±8Coding IQ 106.6, 95% interval 99–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.25cache read $0.125batch $0.625 | $10.00batch $5.00 | 272K | 49.7Fair value (49.7). Costs per task about what its benchmark results predict. Plenty of data behind it. | 50 ±9Typical (50 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 150.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GLM 5.1 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. Newer version | 130.8From 224 benchmarks. Plenty of data behind it. | 110 ±6Coding IQ 110.0, 95% interval 104–116. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.54cache read $0.286 | $4.84 | 203K | 43.2Poor value (43.2). Costs more per task than its benchmark results predict. Plenty of data behind it. | 51 ±8Typical (51 ±8). Generates about as many tokens as models with similar coding results. Measured at default effort. A fair amount of data behind it. | 150.7Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
DeepSeek V4 Pro 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 130.7From 303 benchmarks. Plenty of data behind it. | 105 ±7Coding IQ 104.6, 95% interval 98–111. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.925cache read $0.165 | $3.828 | 1M | 46.9Poor value (46.9). Costs more per task than its benchmark results predict. Plenty of data behind it. | 43 ±10Terse (43 ±10). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 150.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
O3 Pro 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 130.0From 28 benchmarks. Plenty of data behind it. | —Not enough data yet | $20.00batch $10.00 | $80.00batch $40.00 | 200K | 36.5Poor value (36.5). Costs more per task than its benchmark results predict. A fair amount of data behind it. | —Not measured | 147.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
O3 Pro 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 130.0From 28 benchmarks. Plenty of data behind it. | —Not enough data yet | $10.00 | $40.00 | —Not published by the provider | 36.5Poor value (36.5). Costs more per task than its benchmark results predict. A fair amount of data behind it. | —Not measured | 147.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
O3 Pro 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 130.0From 28 benchmarks. Plenty of data behind it. | —Not enough data yet | $20.00batch $10.00 | $80.00batch $40.00 | 200K | 36.5Poor value (36.5). Costs more per task than its benchmark results predict. A fair amount of data behind it. | —Not measured | 147.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Run by Microsoft Azure
58 listings on this page where Microsoft Azure runs the model. Another provider sells these listings.
Select a score for details.
About these columns
- llminfo IQ
- llminfo IQ (capability index). How llminfo IQ is computed
- Coding IQ
- llminfo Coding IQ (coding capability, its own scale). How Coding IQ is computed
- Value
- Value for money. 50 means as expected for its results. Above 50, cheaper per task. How Value for money works
- Yap
- Tokens used per coding task vs. models with the same results. Includes agents that write code. Chattiness on maths doesn't predict it. How the Yap score works
- Epoch Capabilities Index
- Epoch Capabilities Index benchmark details
| Model | Provider | llminfo IQ | Coding IQ | Input $/1M | Output $/1M | Max Input Tokens | Value | Yap | Epoch Capabilities Index |
|---|---|---|---|---|---|---|---|---|---|
Claude Opus 4.6 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 136.9From 466 benchmarks. Plenty of data behind it. | 118 ±5Coding IQ 117.9, 95% interval 113–123. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $5.00cache read $0.50 | $25.00cache write $6.25 | 1M | 47.2Poor value (47.2). Costs more per task than its benchmark results predict. Plenty of data behind it. | 41 ±15Terse (41 ±15). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 156.3Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
136.8From 489 benchmarks. Plenty of data behind it. | 112 ±8Coding IQ 112.0, 95% interval 104–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | — | — | 1.1M | 60.1Good value (60.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 28 ±8Terse (28 ±8). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 157.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
136.8From 489 benchmarks. Plenty of data behind it. | 112 ±8Coding IQ 112.0, 95% interval 104–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.50cache read $0.25 | $15.00 | —Not published by the provider | 60.1Good value (60.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 28 ±8Terse (28 ±8). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 157.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
136.8From 489 benchmarks. Plenty of data behind it. | 112 ±8Coding IQ 112.0, 95% interval 104–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.50cache read $0.25 | $15.00 | 922K | 60.1Good value (60.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 28 ±8Terse (28 ±8). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 157.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
136.8From 489 benchmarks. Plenty of data behind it. | 112 ±8Coding IQ 112.0, 95% interval 104–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.50cache read $0.25 | $15.00 | 1.1M | 60.1Good value (60.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 28 ±8Terse (28 ±8). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 157.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
136.8From 489 benchmarks. Plenty of data behind it. | 112 ±8Coding IQ 112.0, 95% interval 104–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.50cache read $0.25 | $15.00 | 1.1M | 60.1Good value (60.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 28 ±8Terse (28 ±8). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 157.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
136.8From 489 benchmarks. Plenty of data behind it. | 112 ±8Coding IQ 112.0, 95% interval 104–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.75cache read $0.275 | $16.50 | 1.1M | 60.1Good value (60.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 28 ±8Terse (28 ±8). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 157.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
136.8From 489 benchmarks. Plenty of data behind it. | 112 ±8Coding IQ 112.0, 95% interval 104–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.75cache read $0.275 | $16.50 | 1.1M | 60.1Good value (60.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 28 ±8Terse (28 ±8). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 157.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
Claude Sonnet 5 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 136.3From 360 benchmarks. Plenty of data behind it. | 115 ±5Coding IQ 115.2, 95% interval 111–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.00cache read $0.20 | $10.00cache write $2.50 | 1M | 37.4Poor value (37.4). Costs more per task than its benchmark results predict. Plenty of data behind it. | 57 ±9Chatty (57 ±9). Generates more tokens than models with similar coding results. Measured at medium effort. Little data behind it, so treat it as a rough estimate. | 156.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.6 Terra 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 135.7From 329 benchmarks. Plenty of data behind it. | 121 ±5Coding IQ 120.9, 95% interval 116–126. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | — | — | 1.1M | 60.4Good value (60.4). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 48 ±10Typical (48 ±10). Generates about as many tokens as models with similar coding results. Measured at max effort. A fair amount of data behind it. | 160.8Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.6 Terra 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 135.7From 329 benchmarks. Plenty of data behind it. | 121 ±5Coding IQ 120.9, 95% interval 116–126. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.00cache read $0.20 | $12.00cache write $2.50 | —Not published by the provider | 60.4Good value (60.4). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 48 ±10Typical (48 ±10). Generates about as many tokens as models with similar coding results. Measured at max effort. A fair amount of data behind it. | 160.8Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.6 Terra 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 135.7From 329 benchmarks. Plenty of data behind it. | 121 ±5Coding IQ 120.9, 95% interval 116–126. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.00cache read $0.20 | $12.00cache write $2.50 | 922K | 60.4Good value (60.4). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 48 ±10Typical (48 ±10). Generates about as many tokens as models with similar coding results. Measured at max effort. A fair amount of data behind it. | 160.8Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.6 Terra 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 135.7From 329 benchmarks. Plenty of data behind it. | 121 ±5Coding IQ 120.9, 95% interval 116–126. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.00cache read $0.20 | $12.00cache write $2.50 | 922K | 60.4Good value (60.4). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 48 ±10Typical (48 ±10). Generates about as many tokens as models with similar coding results. Measured at max effort. A fair amount of data behind it. | 160.8Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.6 Terra 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 135.7From 329 benchmarks. Plenty of data behind it. | 121 ±5Coding IQ 120.9, 95% interval 116–126. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.00cache read $0.20 | $12.00cache write $2.50 | 1.1M | 60.4Good value (60.4). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 48 ±10Typical (48 ±10). Generates about as many tokens as models with similar coding results. Measured at max effort. A fair amount of data behind it. | 160.8Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.6 Terra 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 135.7From 329 benchmarks. Plenty of data behind it. | 121 ±5Coding IQ 120.9, 95% interval 116–126. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.20cache read $0.22 | $13.20cache write $2.75 | 1.1M | 60.4Good value (60.4). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 48 ±10Typical (48 ±10). Generates about as many tokens as models with similar coding results. Measured at max effort. A fair amount of data behind it. | 160.8Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-6 Luna 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 135.6From 227 benchmarks. Plenty of data behind it. | 117 ±6Coding IQ 116.9, 95% interval 111–123. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | — | — | 1.1M | 63.4Good value (63.4). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 51 ±11Typical (51 ±11). Generates about as many tokens as models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate. | 156.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-6 Luna 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 135.6From 227 benchmarks. Plenty of data behind it. | 117 ±6Coding IQ 116.9, 95% interval 111–123. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.10cache read $0.01 | $0.50cache write $0.125 | —Not published by the provider | 63.4Good value (63.4). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 51 ±11Typical (51 ±11). Generates about as many tokens as models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate. | 156.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-6 Luna 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 135.6From 227 benchmarks. Plenty of data behind it. | 117 ±6Coding IQ 116.9, 95% interval 111–123. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.10cache read $0.01 | $0.50cache write $0.125 | 922K | 63.4Good value (63.4). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 51 ±11Typical (51 ±11). Generates about as many tokens as models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate. | 156.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-6 Luna 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 135.6From 227 benchmarks. Plenty of data behind it. | 117 ±6Coding IQ 116.9, 95% interval 111–123. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.10cache read $0.01 | $0.50cache write $0.125 | 922K | 63.4Good value (63.4). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 51 ±11Typical (51 ±11). Generates about as many tokens as models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate. | 156.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.6 Luna 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 135.2From 344 benchmarks. Plenty of data behind it. | 112 ±10Coding IQ 112.4, 95% interval 102–123. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | — | — | 1.1M | 66.2Good value (66.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 52 ±8Typical (52 ±8). Generates about as many tokens as models with similar coding results. Measured at max effort. A fair amount of data behind it. | 156.7Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.6 Luna 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 135.2From 344 benchmarks. Plenty of data behind it. | 112 ±10Coding IQ 112.4, 95% interval 102–123. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.20cache read $0.02 | $1.20cache write $0.25 | —Not published by the provider | 66.2Good value (66.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 52 ±8Typical (52 ±8). Generates about as many tokens as models with similar coding results. Measured at max effort. A fair amount of data behind it. | 156.7Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.6 Luna 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 135.2From 344 benchmarks. Plenty of data behind it. | 112 ±10Coding IQ 112.4, 95% interval 102–123. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.20cache read $0.02 | $1.20cache write $0.25 | 922K | 66.2Good value (66.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 52 ±8Typical (52 ±8). Generates about as many tokens as models with similar coding results. Measured at max effort. A fair amount of data behind it. | 156.7Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.6 Luna 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 135.2From 344 benchmarks. Plenty of data behind it. | 112 ±10Coding IQ 112.4, 95% interval 102–123. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.20cache read $0.02 | $1.20cache write $0.25 | 922K | 66.2Good value (66.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 52 ±8Typical (52 ±8). Generates about as many tokens as models with similar coding results. Measured at max effort. A fair amount of data behind it. | 156.7Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.6 Luna 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 135.2From 344 benchmarks. Plenty of data behind it. | 112 ±10Coding IQ 112.4, 95% interval 102–123. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.20cache read $0.02 | $1.20cache write $0.25 | 1.1M | 66.2Good value (66.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 52 ±8Typical (52 ±8). Generates about as many tokens as models with similar coding results. Measured at max effort. A fair amount of data behind it. | 156.7Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.6 Luna 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 135.2From 344 benchmarks. Plenty of data behind it. | 112 ±10Coding IQ 112.4, 95% interval 102–123. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.22cache read $0.022 | $1.32cache write $0.275 | 1.1M | 66.2Good value (66.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 52 ±8Typical (52 ±8). Generates about as many tokens as models with similar coding results. Measured at max effort. A fair amount of data behind it. | 156.7Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.2 Pro 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 134.7From 42 benchmarks. Plenty of data behind it. | —Not enough data yet | $21.00 | $168.00 | —Not published by the provider | 49.6Fair value (49.6). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | 157.4Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
Claude Haiku 5.5 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 134.6From 230 benchmarks. Plenty of data behind it. | 124 ±6Coding IQ 123.6, 95% interval 118–129. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.10cache read $0.01 | $0.50cache write $0.125 | 1M | 62.1Good value (62.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 49 ±9Typical (49 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 160.3Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
Claude Opus 4.5 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 134.0From 384 benchmarks. Plenty of data behind it. | 113 ±5Coding IQ 112.8, 95% interval 108–118. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $5.00cache read $0.50 | $25.00cache write $6.25 | 200K | 46.0Poor value (46.0). Costs more per task than its benchmark results predict. Plenty of data behind it. | 43 ±10Terse (43 ±10). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 151.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.1 Codex Max 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 134.0From 11 benchmarks. A fair amount of data behind it. | 112 ±9Coding IQ 111.7, 95% interval 103–121. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.25cache read $0.125 | $10.00 | 272K | 47.0Poor value (47.0). Costs more per task than its benchmark results predict. A fair amount of data behind it. | 48 ±9Typical (48 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | —Not measured | |
GPT-5.1 Codex Max 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 134.0From 11 benchmarks. A fair amount of data behind it. | 112 ±9Coding IQ 111.7, 95% interval 103–121. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.25cache read $0.125 | $10.00 | 272K | 47.0Poor value (47.0). Costs more per task than its benchmark results predict. A fair amount of data behind it. | 48 ±9Typical (48 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | —Not measured | |
Claude Sonnet 4.6 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 133.8From 469 benchmarks. Plenty of data behind it. | 108 ±6Coding IQ 108.4, 95% interval 103–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $3.00cache read $0.30 | $15.00cache write $3.75 | 1M | 35.4Poor value (35.4). Costs more per task than its benchmark results predict. Plenty of data behind it. | 71 ±8Chatty (71 ±8). Generates more tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 152.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.3 Codex 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 133.6From 51 benchmarks. Plenty of data behind it. | 114 ±8Coding IQ 113.7, 95% interval 106–122. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.75cache read $0.175 | $14.00 | —Not published by the provider | 59.7Good value (59.7). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 46 ±9Typical (46 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | —Not measured | |
GPT-5.3 Codex 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 133.6From 51 benchmarks. Plenty of data behind it. | 114 ±8Coding IQ 113.7, 95% interval 106–122. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.75cache read $0.175 | $14.00 | 272K | 59.7Good value (59.7). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 46 ±9Typical (46 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | —Not measured | |
GPT-5.3 Codex 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 133.6From 51 benchmarks. Plenty of data behind it. | 114 ±8Coding IQ 113.7, 95% interval 106–122. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.75cache read $0.175 | $14.00 | 400K | 59.7Good value (59.7). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 46 ±9Typical (46 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | —Not measured | |
133.4From 440 benchmarks. Plenty of data behind it. | 112 ±7Coding IQ 111.5, 95% interval 105–118. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.75cache read $0.175 | $14.00 | —Not published by the provider | 56.6Good value (56.6). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 49 ±14Typical (49 ±14). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 154.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
133.4From 440 benchmarks. Plenty of data behind it. | 112 ±7Coding IQ 111.5, 95% interval 105–118. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.75cache read $0.175 | $14.00 | 272K | 56.6Good value (56.6). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 49 ±14Typical (49 ±14). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 154.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
133.4From 440 benchmarks. Plenty of data behind it. | 112 ±7Coding IQ 111.5, 95% interval 105–118. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.75cache read $0.175 | $14.00 | 400K | 56.6Good value (56.6). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 49 ±14Typical (49 ±14). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 154.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
GPT-5 Pro 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 132.8From 14 benchmarks. Plenty of data behind it. | —Not enough data yet | $15.00 | $120.00 | —Not published by the provider | 29.6Poor value (29.6). Costs more per task than its benchmark results predict. A fair amount of data behind it. | —Not measured | 151.3Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.2 Codex 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 131.9From 52 benchmarks. Plenty of data behind it. | 110 ±6Coding IQ 109.8, 95% interval 104–116. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.75cache read $0.175 | $14.00 | 272K | 52.6Good value (52.6). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 59 ±11Chatty (59 ±11). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | —Not measured | |
GPT-5.2 Codex 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 131.9From 52 benchmarks. Plenty of data behind it. | 110 ±6Coding IQ 109.8, 95% interval 104–116. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.75cache read $0.175 | $14.00 | —Not published by the provider | 52.6Good value (52.6). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 59 ±11Chatty (59 ±11). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | —Not measured | |
GPT-5.2 Codex 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 131.9From 52 benchmarks. Plenty of data behind it. | 110 ±6Coding IQ 109.8, 95% interval 104–116. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.75cache read $0.175 | $14.00 | 272K | 52.6Good value (52.6). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 59 ±11Chatty (59 ±11). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | —Not measured | |
GPT-5.2 Codex 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 131.9From 52 benchmarks. Plenty of data behind it. | 110 ±6Coding IQ 109.8, 95% interval 104–116. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.75cache read $0.175 | $14.00 | 400K | 52.6Good value (52.6). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 59 ±11Chatty (59 ±11). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | —Not measured | |
131.7From 584 benchmarks. Plenty of data behind it. | 111 ±5Coding IQ 110.8, 95% interval 106–116. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | — | — | 400K | 52.5Fair value (52.5). Costs per task about what its benchmark results predict. Plenty of data behind it. | 65 ±8Chatty (65 ±8). Generates more tokens than models with similar coding results. Measured at medium effort. A fair amount of data behind it. | 150Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
131.7From 584 benchmarks. Plenty of data behind it. | 111 ±5Coding IQ 110.8, 95% interval 106–116. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.25cache read $0.125 | $10.00 | —Not published by the provider | 52.5Fair value (52.5). Costs per task about what its benchmark results predict. Plenty of data behind it. | 65 ±8Chatty (65 ±8). Generates more tokens than models with similar coding results. Measured at medium effort. A fair amount of data behind it. | 150Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
131.7From 584 benchmarks. Plenty of data behind it. | 111 ±5Coding IQ 110.8, 95% interval 106–116. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.25cache read $0.125 | $10.00 | 272K | 52.5Fair value (52.5). Costs per task about what its benchmark results predict. Plenty of data behind it. | 65 ±8Chatty (65 ±8). Generates more tokens than models with similar coding results. Measured at medium effort. A fair amount of data behind it. | 150Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
131.7From 584 benchmarks. Plenty of data behind it. | 111 ±5Coding IQ 110.8, 95% interval 106–116. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.25cache read $0.125 | $10.00 | 400K | 52.5Fair value (52.5). Costs per task about what its benchmark results predict. Plenty of data behind it. | 65 ±8Chatty (65 ±8). Generates more tokens than models with similar coding results. Measured at medium effort. A fair amount of data behind it. | 150Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
131.7From 584 benchmarks. Plenty of data behind it. | 111 ±5Coding IQ 110.8, 95% interval 106–116. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.25cache read $0.125 | $10.00 | 400K | 52.5Fair value (52.5). Costs per task about what its benchmark results predict. Plenty of data behind it. | 65 ±8Chatty (65 ±8). Generates more tokens than models with similar coding results. Measured at medium effort. A fair amount of data behind it. | 150Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
131.7From 584 benchmarks. Plenty of data behind it. | 111 ±5Coding IQ 110.8, 95% interval 106–116. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.375cache read $0.1375 | $11.00 | 400K | 52.5Fair value (52.5). Costs per task about what its benchmark results predict. Plenty of data behind it. | 65 ±8Chatty (65 ±8). Generates more tokens than models with similar coding results. Measured at medium effort. A fair amount of data behind it. | 150Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
131.7From 584 benchmarks. Plenty of data behind it. | 111 ±5Coding IQ 110.8, 95% interval 106–116. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.375cache read $0.1375 | $11.00 | 400K | 52.5Fair value (52.5). Costs per task about what its benchmark results predict. Plenty of data behind it. | 65 ±8Chatty (65 ±8). Generates more tokens than models with similar coding results. Measured at medium effort. A fair amount of data behind it. | 150Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
131.0From 352 benchmarks. Plenty of data behind it. | 107 ±8Coding IQ 106.6, 95% interval 99–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | — | — | 400K | 49.7Fair value (49.7). Costs per task about what its benchmark results predict. Plenty of data behind it. | 50 ±9Typical (50 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 150.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
131.0From 352 benchmarks. Plenty of data behind it. | 107 ±8Coding IQ 106.6, 95% interval 99–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.25cache read $0.125 | $10.00 | —Not published by the provider | 49.7Fair value (49.7). Costs per task about what its benchmark results predict. Plenty of data behind it. | 50 ±9Typical (50 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 150.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
131.0From 352 benchmarks. Plenty of data behind it. | 107 ±8Coding IQ 106.6, 95% interval 99–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.25cache read $0.13 | $10.00 | 272K | 49.7Fair value (49.7). Costs per task about what its benchmark results predict. Plenty of data behind it. | 50 ±9Typical (50 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 150.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
131.0From 352 benchmarks. Plenty of data behind it. | 107 ±8Coding IQ 106.6, 95% interval 99–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.25cache read $0.125 | $10.00 | 400K | 49.7Fair value (49.7). Costs per task about what its benchmark results predict. Plenty of data behind it. | 50 ±9Typical (50 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 150.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
131.0From 352 benchmarks. Plenty of data behind it. | 107 ±8Coding IQ 106.6, 95% interval 99–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.25cache read $0.125 | $10.00 | 400K | 49.7Fair value (49.7). Costs per task about what its benchmark results predict. Plenty of data behind it. | 50 ±9Typical (50 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 150.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
131.0From 352 benchmarks. Plenty of data behind it. | 107 ±8Coding IQ 106.6, 95% interval 99–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.375cache read $0.1375 | $11.00 | 400K | 49.7Fair value (49.7). Costs per task about what its benchmark results predict. Plenty of data behind it. | 50 ±9Typical (50 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 150.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
131.0From 352 benchmarks. Plenty of data behind it. | 107 ±8Coding IQ 106.6, 95% interval 99–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.375cache read $0.1375 | $11.00 | 400K | 49.7Fair value (49.7). Costs per task about what its benchmark results predict. Plenty of data behind it. | 50 ±9Typical (50 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 150.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | ||
DeepSeek V4 Pro 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 130.7From 303 benchmarks. Plenty of data behind it. | 105 ±7Coding IQ 104.6, 95% interval 98–111. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.91 | $3.83 | 1M | 46.9Poor value (46.9). Costs more per task than its benchmark results predict. Plenty of data behind it. | 43 ±10Terse (43 ±10). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 150.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
DeepSeek V4 Pro 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 130.7From 303 benchmarks. Plenty of data behind it. | 105 ±7Coding IQ 104.6, 95% interval 98–111. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.91cache read $0.16 | $3.83 | 1M | 46.9Poor value (46.9). Costs more per task than its benchmark results predict. Plenty of data behind it. | 43 ±10Terse (43 ±10). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 150.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence)