Listings
All listings: Microsoft Azure
Every listing for the models on page 3 of 13 of Microsoft Azure, including each model's other versions and routes. Back to Microsoft Azure
Sold by Microsoft Azure
53 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 |
|---|---|---|---|---|---|---|---|---|
Claude Sonnet 4.5 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 129.7From 442 benchmarks. Plenty of data behind it. | 107 ±7Coding IQ 107.2, 95% interval 101–114. 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 | 200K | 40.2Poor value (40.2). Costs more per task than its benchmark results predict. Plenty of data behind it. | 41 ±8Terse (41 ±8). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 147.6Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
129.7From 306 benchmarks. Plenty of data behind it. | 107 ±5Coding IQ 106.7, 95% interval 102–112. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.10cache read $0.22 | $3.52 | 200K | 49.7Fair value (49.7). Costs per task about what its benchmark results predict. Plenty of data behind it. | 55 ±9Typical (55 ±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. | 146.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
Mistral Large 4 🇫🇷Maker headquartered in FR, according to the company's own website. The citation is on the model page. | 129.6From 173 benchmarks. Plenty of data behind it. | 115 ±7Coding IQ 114.7, 95% interval 108–122. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.70 | $2.10 | 1M | 30.8Poor value (30.8). Costs more per task than its benchmark results predict. Plenty of data behind it. | 62 ±9Chatty (62 ±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. | 154.6Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Grok 4.20 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 129.3From 281 benchmarks. Plenty of data behind it. | 108 ±8Coding IQ 108.2, 95% interval 101–116. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.25 | $2.50 | 262K | 40.6Poor value (40.6). Costs more per task than its benchmark results predict. Plenty of data behind it. | 45 ±9Typical (45 ±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. | 152.3Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Grok 4.20 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 129.3From 281 benchmarks. Plenty of data behind it. | 108 ±8Coding IQ 108.2, 95% interval 101–116. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.25 | $2.50 | 262K | 40.6Poor value (40.6). Costs more per task than its benchmark results predict. Plenty of data behind it. | 45 ±9Typical (45 ±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. | 152.3Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
DeepSeek V4 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. Newer version | 129.1From 229 benchmarks. Plenty of data behind it. | 108 ±12Coding IQ 107.8, 95% interval 95–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.15cache read $0.03 | $0.31 | 1M | 74.6Good value (74.6). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 44 ±9Terse (44 ±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 146.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5.1 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 | 129.0From 31 benchmarks. Plenty of data behind it. | 106 ±7Coding IQ 106.2, 95% interval 99–113. 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. | 54 ±9Typical (54 ±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 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 129.0From 31 benchmarks. Plenty of data behind it. | 106 ±7Coding IQ 106.2, 95% interval 99–113. A fair amount 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 | 47.0Poor value (47.0). Costs more per task than its benchmark results predict. A fair amount of data behind it. | 54 ±9Typical (54 ±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 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 129.0From 31 benchmarks. Plenty of data behind it. | 106 ±7Coding IQ 106.2, 95% interval 99–113. 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. | 54 ±9Typical (54 ±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 |
DeepSeek V3.2 Speciale 🇨🇳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. | 128.6From 33 benchmarks. Plenty of data behind it. | —Not enough data yet | $0.58 | $1.68 | 128K | 73.7Good value (73.7). Cheaper per task than its benchmark results predict. Plenty of data behind it. | —Not measured | —Not measured |
Kimi K2.5 🇨🇳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 | 128.5From 401 benchmarks. Plenty of data behind it. | 103 ±5Coding IQ 102.9, 95% interval 98–108. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.66cache read $0.11 | $3.30 | 262K | 48.2Fair value (48.2). Costs per task about what its benchmark results predict. Plenty of data behind it. | 42 ±9Terse (42 ±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 148.8Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
OpenAI o3 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 128.3From 343 benchmarks. Plenty of data behind it. | 100 ±6Coding IQ 99.6, 95% interval 93–106. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.00cache read $0.50batch $1.00 | $8.00batch $4.00 | —Not published by the provider | 54.5Good value (54.5). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 58 ±9Chatty (58 ±9). Generates more tokens than models with similar coding results. Measured at medium effort. A fair amount of data behind it. | 147Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
OpenAI o3 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 128.3From 343 benchmarks. Plenty of data behind it. | 100 ±6Coding IQ 99.6, 95% interval 93–106. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.00cache read $0.50batch $1.00 | $8.00batch $4.00 | 200K | 54.5Good value (54.5). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 58 ±9Chatty (58 ±9). Generates more tokens than models with similar coding results. Measured at medium effort. A fair amount of data behind it. | 147Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
OpenAI o3 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. o3-2025-04-16 · 2025-04-16 · 🇦🇺🇧🇷+17 australia-east, brazil-south +22o3-2025-04-16 · 2025-04-16 · 🇦🇺🇧🇷+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 | 128.3From 343 benchmarks. Plenty of data behind it. | 100 ±6Coding IQ 99.6, 95% interval 93–106. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.00cache read $0.50batch $1.00 | $8.00batch $4.00 | 200K | 54.5Good value (54.5). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 58 ±9Chatty (58 ±9). Generates more tokens than models with similar coding results. Measured at medium effort. A fair amount of data behind it. | 147Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
OpenAI o3 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 128.3From 343 benchmarks. Plenty of data behind it. | 100 ±6Coding IQ 99.6, 95% interval 93–106. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.20cache read $0.55 | $8.80 | —Not published by the provider | 54.5Good value (54.5). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 58 ±9Chatty (58 ±9). Generates more tokens than models with similar coding results. Measured at medium effort. A fair amount of data behind it. | 147Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
OpenAI o3 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 128.3From 343 benchmarks. Plenty of data behind it. | 100 ±6Coding IQ 99.6, 95% interval 93–106. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.20cache read $0.55 | $8.80 | —Not published by the provider | 54.5Good value (54.5). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 58 ±9Chatty (58 ±9). Generates more tokens than models with similar coding results. Measured at medium effort. A fair amount of data behind it. | 147Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
OpenAI o3 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 128.3From 343 benchmarks. Plenty of data behind it. | 100 ±6Coding IQ 99.6, 95% interval 93–106. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.00 | $4.00 | —Not published by the provider | 54.5Good value (54.5). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 58 ±9Chatty (58 ±9). Generates more tokens than models with similar coding results. Measured at medium effort. A fair amount of data behind it. | 147Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5.4 mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. 🇦🇺🇧🇷+18 australia-east, brazil-south +24🇦🇺🇧🇷+18 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, qatar-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 | 128.1From 307 benchmarks. Plenty of data behind it. | 104 ±5Coding IQ 104.3, 95% interval 99–110. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.75cache read $0.075batch $0.375 | $4.50batch $2.25 | 272K | 39.8Poor value (39.8). Costs more per task than its benchmark results predict. Plenty of data behind it. | 46 ±18Typical (46 ±18). 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. | 149.4Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5.4 mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 128.1From 307 benchmarks. Plenty of data behind it. | 104 ±5Coding IQ 104.3, 95% interval 99–110. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.825cache read $0.0825 | $4.95 | —Not published by the provider | 39.8Poor value (39.8). Costs more per task than its benchmark results predict. Plenty of data behind it. | 46 ±18Typical (46 ±18). 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. | 149.4Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5.4 mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 128.1From 307 benchmarks. Plenty of data behind it. | 104 ±5Coding IQ 104.3, 95% interval 99–110. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.75cache read $0.075batch $0.375 | $4.50batch $2.25 | 272K | 39.8Poor value (39.8). Costs more per task than its benchmark results predict. Plenty of data behind it. | 46 ±18Typical (46 ±18). 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. | 149.4Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5 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 +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 | 127.7From 21 benchmarks. Plenty of data behind it. | ≈ 106 ±9staleCoding IQ 106.4, 95% interval 98–115. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ. Its newest result is over a year old. | $1.25cache read $0.125 | $10.00 | 272K | 49.4Fair value (49.4). Costs per task about what its benchmark results predict. A fair amount of data behind it. | —Not measured | —Not measured |
Inkling FP4 🇺🇸Maker headquartered in US, 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. | 126.9From 215 benchmarks. Plenty of data behind it. | 102 ±6Coding IQ 101.9, 95% interval 96–107. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.10cache read $0.19 | $4.46 | 1M | 27.9Poor value (27.9). Costs more per task than its benchmark results predict. Plenty of data behind it. | 58 ±9Chatty (58 ±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. | 148.4Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Kimi K2.7 Code 🇨🇳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. | 126.9From 102 benchmarks. Plenty of data behind it. | 105 ±9Coding IQ 104.5, 95% interval 96–113. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.05cache read $0.21 | $4.40 | 262K | 43.6Poor value (43.6). Costs more per task than its benchmark results predict. Plenty of data behind it. | 62 ±9Chatty (62 ±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. | 151.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
MiniMax M3 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. | 126.8From 332 benchmarks. Plenty of data behind it. | 107 ±9Coding IQ 107.2, 95% interval 98–117. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.33cache read $0.066 | $1.32 | 512K | 50.9Fair value (50.9). Costs per task about what 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. | 147.8Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
O1 Preview 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 125.5From 80 benchmarks. Plenty of data behind it. | ≈ 87 ±15Coding IQ 86.5, 95% interval 72–102. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $15.00cache read $7.50 | $60.00 | —Not published by the provider | 37.5Poor value (37.5). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | 133.6Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
O1 Preview 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. o1-preview-2024-09-12 · 2024-09-12 · 🇺🇸🇮🇪+1 us-east-2, europe-north +1o1-preview-2024-09-12 · 2024-09-12 · 🇺🇸🇮🇪+1 us-east-2, europe-north, sweden-central | 125.5From 80 benchmarks. Plenty of data behind it. | ≈ 87 ±15Coding IQ 86.5, 95% interval 72–102. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $15.00cache read $7.50 | $60.00 | —Not published by the provider | 37.5Poor value (37.5). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | 133.6Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
O1 Preview 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 125.5From 80 benchmarks. Plenty of data behind it. | ≈ 87 ±15Coding IQ 86.5, 95% interval 72–102. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $16.50cache read $8.25 | $66.00 | —Not published by the provider | 37.5Poor value (37.5). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | 133.6Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
O1 Preview 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 125.5From 80 benchmarks. Plenty of data behind it. | ≈ 87 ±15Coding IQ 86.5, 95% interval 72–102. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $16.50cache read $8.25 | $66.00 | —Not published by the provider | 37.5Poor value (37.5). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | 133.6Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
O1 Preview 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 125.5From 80 benchmarks. Plenty of data behind it. | ≈ 87 ±15Coding IQ 86.5, 95% interval 72–102. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $7.50 | $30.00 | —Not published by the provider | 37.5Poor value (37.5). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | 133.6Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
O1 Preview 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 125.5From 80 benchmarks. Plenty of data behind it. | ≈ 87 ±15Coding IQ 86.5, 95% interval 72–102. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $7.50 | $30.00 | —Not published by the provider | 37.5Poor value (37.5). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | 133.6Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Grok 4 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 125.2From 296 benchmarks. Plenty of data behind it. | 98 ±13Coding IQ 98.0, 95% interval 85–111. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $3.00 | $15.00 | 262K | 32.7Poor value (32.7). Costs more per task than its benchmark results predict. Plenty of data behind it. | —Not measured | 146.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
DeepSeek V3.2 🇨🇳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 | 125.0From 247 benchmarks. Plenty of data behind it. | 101 ±7Coding IQ 101.2, 95% interval 94–108. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.62cache read $0.31 | $1.85 | 164K | 63.8Good value (63.8). Cheaper per task than its benchmark results predict. Plenty of data behind it. | —Not measured | 146.3Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Mai Thinking 1 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 124.8From 17 benchmarks. A fair amount of data behind it. | —Not enough data yet | $2.00cache read $0.20 | $8.00 | 256K | —Not enough data yet | —Not measured | —Not measured |
GPT-5 mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version🇦🇺🇧🇷+19 australia-east, brazil-south +26🇦🇺🇧🇷+19 australia-east, brazil-south, canada-east, us-central, us-east, us-east-2, france-central, germany-west-central, italy-north, japan-east, korea-central, us-north-central, europe-north, norway-east, poland-central, qatar-central, south-africa-north, us-south-central, asia-pacific-southeast, south-india, spain-central, sweden-central, switzerland-north, uae-north, united-kingdom-south, europe-west, us-west, us-west-3 | 124.6From 388 benchmarks. Plenty of data behind it. | 98 ±14Coding IQ 98.3, 95% interval 85–112. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.25cache read $0.025batch $0.125 | $2.00batch $1.00 | 272K | 55.1Good value (55.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 40 ±9Terse (40 ±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 146Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5 mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 124.6From 388 benchmarks. Plenty of data behind it. | 98 ±14Coding IQ 98.3, 95% interval 85–112. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.275cache read $0.0275 | $2.20 | —Not published by the provider | 55.1Good value (55.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 40 ±9Terse (40 ±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 146Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5 mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 124.6From 388 benchmarks. Plenty of data behind it. | 98 ±14Coding IQ 98.3, 95% interval 85–112. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.25cache read $0.0275batch $0.1375 | $2.00batch $1.10 | 272K | 55.1Good value (55.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 40 ±9Terse (40 ±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 146Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5 mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 124.6From 388 benchmarks. Plenty of data behind it. | 98 ±14Coding IQ 98.3, 95% interval 85–112. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.125 | $1.00 | —Not published by the provider | 55.1Good value (55.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 40 ±9Terse (40 ±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 146Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5 mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 124.6From 388 benchmarks. Plenty of data behind it. | 98 ±14Coding IQ 98.3, 95% interval 85–112. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.125 | $1.00 | —Not published by the provider | 55.1Good value (55.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 40 ±9Terse (40 ±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 146Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
124.4From 11 benchmarks. Plenty of data behind it. | —Not enough data yet | $150.00cache read $75.00batch $75.00 | $600.00batch $300.00 | —Not published by the provider | 39.3Poor value (39.3). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | —Not measured | |
124.4From 11 benchmarks. Plenty of data behind it. | —Not enough data yet | $75.00 | $300.00 | —Not published by the provider | 39.3Poor value (39.3). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | —Not measured | |
Grok 4.3 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 124.2From 205 benchmarks. Plenty of data behind it. | 101 ±8Coding IQ 101.2, 95% interval 94–109. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.25>200k $2.50cache read $0.20 | $2.50>200k $5.00 | 200K | 39.3Poor value (39.3). Costs more per task than its benchmark results predict. Plenty 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 |
Nemotron 3 Ultra 🇺🇸Maker headquartered in US, 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. | 124.1From 172 benchmarks. Plenty of data behind it. | 101 ±12Coding IQ 100.9, 95% interval 89–112. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.66cache read $0.13 | $2.64 | 262K | 33.1Poor value (33.1). Costs more per task than its benchmark results predict. A fair amount of data behind it. | 53 ±9Typical (53 ±9). Generates about as many tokens as models with similar coding results. Measured with reasoning off. Little data behind it, so treat it as a rough estimate. | 147Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
o4 Mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 124.0From 384 benchmarks. Plenty of data behind it. | 101 ±14Coding IQ 100.7, 95% interval 87–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.10cache read $0.275batch $0.55 | $4.40batch $2.20 | —Not published by the provider | 56.2Good value (56.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 53 ±16Typical (53 ±16). Generates about as many tokens as models with similar coding results. Measured at low effort. A fair amount of data behind it. | 145.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
o4 Mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 124.0From 384 benchmarks. Plenty of data behind it. | 101 ±14Coding IQ 100.7, 95% interval 87–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.10cache read $0.28batch $0.55 | $4.40batch $2.20 | 200K | 56.2Good value (56.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 53 ±16Typical (53 ±16). Generates about as many tokens as models with similar coding results. Measured at low effort. A fair amount of data behind it. | 145.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
o4 Mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. o4-mini-2025-04-16 · 2025-04-16 · 🇦🇺🇧🇷+17 australia-east, brazil-south +24o4-mini-2025-04-16 · 2025-04-16 · 🇦🇺🇧🇷+17 australia-east, brazil-south, canada-east, us-central, us-east, us-east-2, france-central, germany-west-central, italy-north, japan-east, korea-central, us-north-central, norway-east, poland-central, south-africa-north, us-south-central, asia-pacific-southeast, south-india, spain-central, sweden-central, switzerland-north, uae-north, united-kingdom-south, europe-west, us-west, us-west-3 | 124.0From 384 benchmarks. Plenty of data behind it. | 101 ±14Coding IQ 100.7, 95% interval 87–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.10cache read $0.275batch $0.55 | $4.40batch $2.20 | 200K | 56.2Good value (56.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 53 ±16Typical (53 ±16). Generates about as many tokens as models with similar coding results. Measured at low effort. A fair amount of data behind it. | 145.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
o4 Mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 124.0From 384 benchmarks. Plenty of data behind it. | 101 ±14Coding IQ 100.7, 95% interval 87–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.21cache read $0.303 | $4.84 | —Not published by the provider | 56.2Good value (56.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 53 ±16Typical (53 ±16). Generates about as many tokens as models with similar coding results. Measured at low effort. A fair amount of data behind it. | 145.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
o4 Mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 124.0From 384 benchmarks. Plenty of data behind it. | 101 ±14Coding IQ 100.7, 95% interval 87–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.21cache read $0.303 | $4.84 | —Not published by the provider | 56.2Good value (56.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 53 ±16Typical (53 ±16). Generates about as many tokens as models with similar coding results. Measured at low effort. A fair amount of data behind it. | 145.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
o4 Mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 124.0From 384 benchmarks. Plenty of data behind it. | 101 ±14Coding IQ 100.7, 95% interval 87–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.55 | $2.20 | —Not published by the provider | 56.2Good value (56.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 53 ±16Typical (53 ±16). Generates about as many tokens as models with similar coding results. Measured at low effort. A fair amount of data behind it. | 145.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
o4 Mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 124.0From 384 benchmarks. Plenty of data behind it. | 101 ±14Coding IQ 100.7, 95% interval 87–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.55 | $2.20 | —Not published by the provider | 56.2Good value (56.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 53 ±16Typical (53 ±16). Generates about as many tokens as models with similar coding results. Measured at low effort. A fair amount of data behind it. | 145.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5.1 Codex Mini 🇺🇸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 | 123.8From 24 benchmarks. Plenty of data behind it. | —Not enough data yet | $0.25cache read $0.025 | $2.00 | 272K | 62.1Good value (62.1). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | —Not measured |
GPT-5.1 Codex Mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 123.8From 24 benchmarks. Plenty of data behind it. | —Not enough data yet | $0.125 | $1.00 | —Not published by the provider | 62.1Good value (62.1). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | —Not measured |
GPT-5.1 Codex Mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 123.8From 24 benchmarks. Plenty of data behind it. | —Not enough data yet | $0.25cache read $0.025 | $2.00 | 272K | 62.1Good value (62.1). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | —Not measured |
Kimi K2 Thinking 🇨🇳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. | 123.8From 184 benchmarks. Plenty of data behind it. | 100 ±6Coding IQ 100.2, 95% interval 94–106. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.60 | $2.50 | 262K | 41.6Poor value (41.6). Costs more per task than its benchmark results predict. Plenty of data behind it. | 50 ±18Typical (50 ±18). 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. | 145.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Run by Microsoft Azure
26 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 Sonnet 4.5 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 129.7From 442 benchmarks. Plenty of data behind it. | 107 ±7Coding IQ 107.2, 95% interval 101–114. 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 | 200K | 40.2Poor value (40.2). Costs more per task than its benchmark results predict. Plenty of data behind it. | 41 ±8Terse (41 ±8). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 147.6Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
DeepSeek V4 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. Newer version | 129.1From 229 benchmarks. Plenty of data behind it. | 108 ±12Coding IQ 107.8, 95% interval 95–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.21 | $0.56 | 1M | 74.6Good value (74.6). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 44 ±9Terse (44 ±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 146.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
DeepSeek V4 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. Newer version | 129.1From 229 benchmarks. Plenty of data behind it. | 108 ±12Coding IQ 107.8, 95% interval 95–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.21cache read $0.031 | $0.56 | 1M | 74.6Good value (74.6). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 44 ±9Terse (44 ±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 146.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.1 Codex 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 129.0From 31 benchmarks. Plenty of data behind it. | 106 ±7Coding IQ 106.2, 95% interval 99–113. 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. | 54 ±9Typical (54 ±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 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 129.0From 31 benchmarks. Plenty of data behind it. | 106 ±7Coding IQ 106.2, 95% interval 99–113. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.25 | $10.00 | —Not published by the provider | 47.0Poor value (47.0). Costs more per task than its benchmark results predict. A fair amount of data behind it. | 54 ±9Typical (54 ±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 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 129.0From 31 benchmarks. Plenty of data behind it. | 106 ±7Coding IQ 106.2, 95% interval 99–113. 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 $0.13 | $10.00 | 272K | 47.0Poor value (47.0). Costs more per task than its benchmark results predict. A fair amount of data behind it. | 54 ±9Typical (54 ±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 | |
OpenAI o3 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 128.3From 343 benchmarks. Plenty of data behind it. | 100 ±6Coding IQ 99.6, 95% interval 93–106. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $2.00cache read $0.50 | $8.00 | —Not published by the provider | 54.5Good value (54.5). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 58 ±9Chatty (58 ±9). Generates more tokens than models with similar coding results. Measured at medium effort. A fair amount of data behind it. | 147Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.4 mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 128.1From 307 benchmarks. Plenty of data behind it. | 104 ±5Coding IQ 104.3, 95% interval 99–110. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.75cache read $0.075 | $4.50 | —Not published by the provider | 39.8Poor value (39.8). Costs more per task than its benchmark results predict. Plenty of data behind it. | 46 ±18Typical (46 ±18). 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. | 149.4Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.4 mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 128.1From 307 benchmarks. Plenty of data behind it. | 104 ±5Coding IQ 104.3, 95% interval 99–110. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.75cache read $0.075 | $4.50 | 400K | 39.8Poor value (39.8). Costs more per task than its benchmark results predict. Plenty of data behind it. | 46 ±18Typical (46 ±18). 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. | 149.4Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.4 mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 128.1From 307 benchmarks. Plenty of data behind it. | 104 ±5Coding IQ 104.3, 95% interval 99–110. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.75cache read $0.075 | $4.50 | 400K | 39.8Poor value (39.8). Costs more per task than its benchmark results predict. Plenty of data behind it. | 46 ±18Typical (46 ±18). 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. | 149.4Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.4 mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 128.1From 307 benchmarks. Plenty of data behind it. | 104 ±5Coding IQ 104.3, 95% interval 99–110. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.75cache read $0.075 | $4.50 | 400K | 39.8Poor value (39.8). Costs more per task than its benchmark results predict. Plenty of data behind it. | 46 ±18Typical (46 ±18). 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. | 149.4Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5 mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 124.6From 388 benchmarks. Plenty of data behind it. | 98 ±14Coding IQ 98.3, 95% interval 85–112. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | — | — | 400K | 55.1Good value (55.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 40 ±9Terse (40 ±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 146Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5 mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 124.6From 388 benchmarks. Plenty of data behind it. | 98 ±14Coding IQ 98.3, 95% interval 85–112. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.25cache read $0.025 | $2.00 | —Not published by the provider | 55.1Good value (55.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 40 ±9Terse (40 ±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 146Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5 mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 124.6From 388 benchmarks. Plenty of data behind it. | 98 ±14Coding IQ 98.3, 95% interval 85–112. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.25cache read $0.03 | $2.00 | 400K | 55.1Good value (55.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 40 ±9Terse (40 ±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 146Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5 mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 124.6From 388 benchmarks. Plenty of data behind it. | 98 ±14Coding IQ 98.3, 95% interval 85–112. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.25cache read $0.025 | $2.00 | 200K | 55.1Good value (55.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 40 ±9Terse (40 ±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 146Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5 mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 124.6From 388 benchmarks. Plenty of data behind it. | 98 ±14Coding IQ 98.3, 95% interval 85–112. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.25cache read $0.025 | $2.00 | 200K | 55.1Good value (55.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 40 ±9Terse (40 ±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 146Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5 mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 124.6From 388 benchmarks. Plenty of data behind it. | 98 ±14Coding IQ 98.3, 95% interval 85–112. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.275cache read $0.0275 | $2.20 | 200K | 55.1Good value (55.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 40 ±9Terse (40 ±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 146Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5 mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 124.6From 388 benchmarks. Plenty of data behind it. | 98 ±14Coding IQ 98.3, 95% interval 85–112. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.275cache read $0.0275 | $2.20 | 200K | 55.1Good value (55.1). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 40 ±9Terse (40 ±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 146Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
Grok 4.3 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 124.2From 205 benchmarks. Plenty of data behind it. | 101 ±8Coding IQ 101.2, 95% interval 94–109. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.25cache read $0.20 | $2.50 | —Not published by the provider | 39.3Poor value (39.3). Costs more per task than its benchmark results predict. Plenty 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 | |
o4 Mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 124.0From 384 benchmarks. Plenty of data behind it. | 101 ±14Coding IQ 100.7, 95% interval 87–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.10cache read $0.275 | $4.40 | 200K | 56.2Good value (56.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 53 ±16Typical (53 ±16). Generates about as many tokens as models with similar coding results. Measured at low effort. A fair amount of data behind it. | 145.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
o4 Mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 124.0From 384 benchmarks. Plenty of data behind it. | 101 ±14Coding IQ 100.7, 95% interval 87–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.21cache read $0.3025 | $4.84 | 200K | 56.2Good value (56.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 53 ±16Typical (53 ±16). Generates about as many tokens as models with similar coding results. Measured at low effort. A fair amount of data behind it. | 145.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
o4 Mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 124.0From 384 benchmarks. Plenty of data behind it. | 101 ±14Coding IQ 100.7, 95% interval 87–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.21cache read $0.3025 | $4.84 | 200K | 56.2Good value (56.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 53 ±16Typical (53 ±16). Generates about as many tokens as models with similar coding results. Measured at low effort. A fair amount of data behind it. | 145.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
o4 Mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 124.0From 384 benchmarks. Plenty of data behind it. | 101 ±14Coding IQ 100.7, 95% interval 87–114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.10cache read $0.275 | $4.40 | 200K | 56.2Good value (56.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 53 ±16Typical (53 ±16). Generates about as many tokens as models with similar coding results. Measured at low effort. A fair amount of data behind it. | 145.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GPT-5.1 Codex Mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 123.8From 24 benchmarks. Plenty of data behind it. | —Not enough data yet | $0.25cache read $0.025 | $2.00 | 272K | 62.1Good value (62.1). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | —Not measured | |
GPT-5.1 Codex Mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 123.8From 24 benchmarks. Plenty of data behind it. | —Not enough data yet | $0.25cache read $0.025 | $2.00 | —Not published by the provider | 62.1Good value (62.1). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | —Not measured | |
GPT-5.1 Codex Mini 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. | 123.8From 24 benchmarks. Plenty of data behind it. | —Not enough data yet | $0.25cache read $0.03 | $2.00 | 272K | 62.1Good value (62.1). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | —Not measured |
Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence)