Listings
All listings: OpenRouter
Every listing for the models on page 8 of 26 of OpenRouter, including each model's other versions and routes. Back to OpenRouter
Sold by OpenRouter
73 listings on this page sold by OpenRouter.
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 |
|---|---|---|---|---|---|---|---|---|
Gemini 2.5 Flash πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 121.6From 327 benchmarks. Plenty of data behind it. | 86Β Β±16Coding IQ 85.5, 95% interval 70β102. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.30cache read $0.03 | $2.50 | 1M | 39.0Poor value (39.0). Costs more per task than its benchmark results predict. Plenty of data behind it. | 57 Β±9Chatty (57 Β±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. | 141.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Gemini 2.5 Flash Inference by GoogleπΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 121.6From 327 benchmarks. Plenty of data behind it. | 86Β Β±16Coding IQ 85.5, 95% interval 70β102. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.30cache read $0.03 | $2.50 | 1M | 39.0Poor value (39.0). Costs more per task than its benchmark results predict. Plenty of data behind it. | 57 Β±9Chatty (57 Β±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. | 141.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Gemini 2.5 Flash Inference by GoogleπΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 121.6From 327 benchmarks. Plenty of data behind it. | 86Β Β±16Coding IQ 85.5, 95% interval 70β102. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.15cache read $0.015 | $1.25 | 1M | 39.0Poor value (39.0). Costs more per task than its benchmark results predict. Plenty of data behind it. | 57 Β±9Chatty (57 Β±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. | 141.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Gemini 2.5 Flash Inference by GoogleπΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 121.6From 327 benchmarks. Plenty of data behind it. | 86Β Β±16Coding IQ 85.5, 95% interval 70β102. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.54cache read $0.054 | $4.50 | 1M | 39.0Poor value (39.0). Costs more per task than its benchmark results predict. Plenty of data behind it. | 57 Β±9Chatty (57 Β±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. | 141.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
DeepSeek V3.1 Terminus π¨π³Maker headquartered in CN, according to the company's own website. The citation is on the model page. | 121.4From 57 benchmarks. Plenty of data behind it. | β 99Β Β±13Coding IQ 99.4, 95% interval 86β112. 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. | $0.27 | $1.00 | 164K | 57.6Good value (57.6). Cheaper per task than its benchmark results predict. A fair amount of data behind it. | 50 Β±9Typical (50 Β±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. | βNot measured |
DeepSeek V3.1 Terminus Inference by SiliconFlowπ¨π³Maker headquartered in CN, according to the company's own website. The citation is on the model page. | 121.4From 57 benchmarks. Plenty of data behind it. | β 99Β Β±13Coding IQ 99.4, 95% interval 86β112. 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. | $0.27 | $1.00 | 164K | 57.6Good value (57.6). Cheaper per task than its benchmark results predict. A fair amount of data behind it. | 50 Β±9Typical (50 Β±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. | βNot measured |
Gemini 2.5 Flash Preview 05-20 (no think) πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | 121.2From 60 benchmarks. Plenty of data behind it. | βNot enough data yet | $0.15cache read $0.0375 | $3.50cache write $0.2333 | βNot published by the provider | 38.2Poor value (38.2). Costs more per task than its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
Qwen3.5 35B A3B π¨π³Maker 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 | 121.1From 311 benchmarks. Plenty of data behind it. | 95Β Β±8Coding IQ 95.0, 95% interval 87β103. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.14cache read $0.05 | $1.00 | 262K | 40.3Poor value (40.3). Costs more per task than its benchmark results predict. Plenty of data behind it. | 59 Β±9Chatty (59 Β±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. | 144.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3.5 35B A3B Inference by Alibaba Cloudπ¨π³Maker 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 | 121.1From 311 benchmarks. Plenty of data behind it. | 95Β Β±8Coding IQ 95.0, 95% interval 87β103. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.1625 | $1.30 | 258K | 40.3Poor value (40.3). Costs more per task than its benchmark results predict. Plenty of data behind it. | 59 Β±9Chatty (59 Β±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. | 144.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3.5 35B A3B Inference by Darkbloomπ¨π³Maker 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 | 121.1From 311 benchmarks. Plenty of data behind it. | 95Β Β±8Coding IQ 95.0, 95% interval 87β103. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.065cache read $0.0325 | $0.75 | 262K | 40.3Poor value (40.3). Costs more per task than its benchmark results predict. Plenty of data behind it. | 59 Β±9Chatty (59 Β±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. | 144.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3.5 35B A3B Inference by DeepInfraπ¨π³Maker 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 | 121.1From 311 benchmarks. Plenty of data behind it. | 95Β Β±8Coding IQ 95.0, 95% interval 87β103. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.14cache read $0.05 | $1.00 | 262K | 40.3Poor value (40.3). Costs more per task than its benchmark results predict. Plenty of data behind it. | 59 Β±9Chatty (59 Β±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. | 144.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3.5 35B A3B Inference by DeepInfraπ¨π³Maker 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 | 121.1From 311 benchmarks. Plenty of data behind it. | 95Β Β±8Coding IQ 95.0, 95% interval 87β103. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.112cache read $0.04 | $0.80 | 262K | 40.3Poor value (40.3). Costs more per task than its benchmark results predict. Plenty of data behind it. | 59 Β±9Chatty (59 Β±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. | 144.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3.5 35B A3B Inference by Parasailπ¨π³Maker 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 | 121.1From 311 benchmarks. Plenty of data behind it. | 95Β Β±8Coding IQ 95.0, 95% interval 87β103. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.15cache read $0.05 | $1.00 | 262K | 40.3Poor value (40.3). Costs more per task than its benchmark results predict. Plenty of data behind it. | 59 Β±9Chatty (59 Β±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. | 144.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3.5 35B A3B Inference by SiliconFlowπ¨π³Maker 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 | 121.1From 311 benchmarks. Plenty of data behind it. | 95Β Β±8Coding IQ 95.0, 95% interval 87β103. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.24cache read $0.15 | $1.80 | 262K | 40.3Poor value (40.3). Costs more per task than its benchmark results predict. Plenty of data behind it. | 59 Β±9Chatty (59 Β±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. | 144.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3.5 35B A3B Inference by Veniceπ¨π³Maker 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 | 121.1From 311 benchmarks. Plenty of data behind it. | 95Β Β±8Coding IQ 95.0, 95% interval 87β103. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.15cache read $0.05 | $1.00 | 256K | 40.3Poor value (40.3). Costs more per task than its benchmark results predict. Plenty of data behind it. | 59 Β±9Chatty (59 Β±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. | 144.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
ERNIE 4.5 VL 424B A47B π¨π³Maker 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. | β 121From 3 benchmarks. Little data behind it, so treat it as a rough estimate. | βNot enough data yet | $0.42 | $1.25 | βNot published by the provider | 54.2Good value (54.2). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Gemini 2.5 Flash Preview πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | 120.8From 67 benchmarks. A fair amount of data behind it. | βNot enough data yet | $0.15cache read $0.0375 | $0.60cache write $0.2333 | βNot published by the provider | βNot enough data yet | βNot measured | βNot measured |
Gemini 2.5 Flash Preview πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | 120.8From 67 benchmarks. A fair amount of data behind it. | βNot enough data yet | $0.15cache read $0.0375 | $3.50cache write $0.2333 | βNot published by the provider | βNot enough data yet | βNot measured | βNot measured |
120.7From 486 benchmarks. Plenty of data behind it. | 92Β Β±6Coding IQ 92.4, 95% interval 87β98. 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 | 1M | 42.8Poor value (42.8). Costs more per task than its benchmark results predict. Plenty of data behind it. | 33 Β±8Terse (33 Β±8). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 136.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
120.7From 486 benchmarks. Plenty of data behind it. | 92Β Β±6Coding IQ 92.4, 95% interval 87β98. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $1.00cache read $0.25 | $4.00 | 1M | 42.8Poor value (42.8). Costs more per task than its benchmark results predict. Plenty of data behind it. | 33 Β±8Terse (33 Β±8). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 136.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
120.7From 486 benchmarks. Plenty of data behind it. | 92Β Β±6Coding IQ 92.4, 95% interval 87β98. 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 | 1M | 42.8Poor value (42.8). Costs more per task than its benchmark results predict. Plenty of data behind it. | 33 Β±8Terse (33 Β±8). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 136.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
120.7From 486 benchmarks. Plenty of data behind it. | 92Β Β±6Coding IQ 92.4, 95% interval 87β98. 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 | 1M | 42.8Poor value (42.8). Costs more per task than its benchmark results predict. Plenty of data behind it. | 33 Β±8Terse (33 Β±8). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it. | 136.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
Qwen3 235B A22B Instruct 2507 π¨π³Maker 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. | 120.7From 258 benchmarks. Plenty of data behind it. | 92Β Β±11Coding IQ 92.4, 95% interval 82β103. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.14cache read $0.05 | $0.80 | 131K | 58.8Good value (58.8). Cheaper per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 138Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3 235B A22B Instruct 2507 Inference by DeepInfraπ¨π³Maker 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. | 120.7From 258 benchmarks. Plenty of data behind it. | 92Β Β±11Coding IQ 92.4, 95% interval 82β103. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.072 | $0.44 | 262K | 58.8Good value (58.8). Cheaper per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 138Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3 235B A22B Instruct 2507 Inference by Parasailπ¨π³Maker 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. | 120.7From 258 benchmarks. Plenty of data behind it. | 92Β Β±11Coding IQ 92.4, 95% interval 82β103. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.14cache read $0.05 | $0.80 | 131K | 58.8Good value (58.8). Cheaper per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 138Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3 235B A22B Instruct 2507 Inference by Veniceπ¨π³Maker 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. | 120.7From 258 benchmarks. Plenty of data behind it. | 92Β Β±11Coding IQ 92.4, 95% interval 82β103. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.15 | $0.75 | 128K | 58.8Good value (58.8). Cheaper per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 138Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Grok 3 Mini πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | 120.4From 190 benchmarks. Plenty of data behind it. | β 95Β Β±12Coding IQ 94.5, 95% interval 82β107. 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. | $0.30cache read $0.075 | $0.50 | βNot published by the provider | 50.6Fair value (50.6). Costs per task about what its benchmark results predict. Plenty of data behind it. | βNot measured | 139.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Grok 3 Mini πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | 120.4From 190 benchmarks. Plenty of data behind it. | β 95Β Β±12Coding IQ 94.5, 95% interval 82β107. 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. | $0.30 | $0.50 | βNot published by the provider | 50.6Fair value (50.6). Costs per task about what its benchmark results predict. Plenty of data behind it. | βNot measured | 139.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3 VL 235B π¨π³Maker 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. | 120.4From 170 benchmarks. Plenty of data behind it. | β 94Β Β±18Coding IQ 94.2, 95% interval 76β113. 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. | $0.26 | $2.60 | βNot published by the provider | 45.0Poor value (45.0). Costs more per task than its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
Qwen3 VL 235B π¨π³Maker 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. | 120.4From 170 benchmarks. Plenty of data behind it. | β 94Β Β±18Coding IQ 94.2, 95% interval 76β113. 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. | $0.21cache read $0.10 | $1.90 | 131K | 45.0Poor value (45.0). Costs more per task than its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
Qwen3 VL 235B Inference by DeepInfraπ¨π³Maker 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. | 120.4From 170 benchmarks. Plenty of data behind it. | β 94Β Β±18Coding IQ 94.2, 95% interval 76β113. 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. | $0.20cache read $0.11 | $0.88 | 262K | 45.0Poor value (45.0). Costs more per task than its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
Qwen3 VL 235B Inference by Parasailπ¨π³Maker 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. | 120.4From 170 benchmarks. Plenty of data behind it. | β 94Β Β±18Coding IQ 94.2, 95% interval 76β113. 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. | $0.21cache read $0.10 | $1.90 | 131K | 45.0Poor value (45.0). Costs more per task than its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
Qwen3 VL 235B Inference by Veniceπ¨π³Maker 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. | 120.4From 170 benchmarks. Plenty of data behind it. | β 94Β Β±18Coding IQ 94.2, 95% interval 76β113. 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. | $0.21cache read $0.10 | $1.90 | 128K | 45.0Poor value (45.0). Costs more per task than its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
GPT 5.4 Nano πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | 120.3From 252 benchmarks. Plenty of data behind it. | 104Β Β±6Coding IQ 103.6, 95% interval 97β110. 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 | $1.25 | 400K | 49.5Fair value (49.5). Costs per task about what its benchmark results predict. Plenty of data behind it. | 47 Β±9Typical (47 Β±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. | 146.4Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT 5.4 Nano πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | 120.3From 252 benchmarks. Plenty of data behind it. | 104Β Β±6Coding IQ 103.6, 95% interval 97β110. 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 | $0.625 | 400K | 49.5Fair value (49.5). Costs per task about what its benchmark results predict. Plenty of data behind it. | 47 Β±9Typical (47 Β±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. | 146.4Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT 5.4 Nano Inference by Microsoft AzureπΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | 120.3From 252 benchmarks. Plenty of data behind it. | 104Β Β±6Coding IQ 103.6, 95% interval 97β110. 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 | $1.25 | 400K | 49.5Fair value (49.5). Costs per task about what its benchmark results predict. Plenty of data behind it. | 47 Β±9Typical (47 Β±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. | 146.4Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT 5.4 Nano Inference by OpenAIπΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | 120.3From 252 benchmarks. Plenty of data behind it. | 104Β Β±6Coding IQ 103.6, 95% interval 97β110. 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 | $1.25 | 400K | 49.5Fair value (49.5). Costs per task about what its benchmark results predict. Plenty of data behind it. | 47 Β±9Typical (47 Β±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. | 146.4Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT 5.4 Nano Inference by OpenAIπΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | 120.3From 252 benchmarks. Plenty of data behind it. | 104Β Β±6Coding IQ 103.6, 95% interval 97β110. 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 | $0.625 | 400K | 49.5Fair value (49.5). Costs per task about what its benchmark results predict. Plenty of data behind it. | 47 Β±9Typical (47 Β±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. | 146.4Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3 Coder Plus π¨π³Maker 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. | 120.2From 11 benchmarks. A fair amount of data behind it. | βNot enough data yet | $0.65cache read $0.13 | $3.25cache write $0.8125 | βNot published by the provider | 46.0Poor value (46.0). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | 39 Β±9Terse (39 Β±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. | βNot measured |
Qwen3 Next 80B A3B π¨π³Maker 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. | 120.2From 191 benchmarks. Plenty of data behind it. | 90Β Β±12Coding IQ 90.0, 95% interval 78β102. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.0975 | $0.78 | βNot published by the provider | 51.1Fair value (51.1). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Step 3.7 Flash π¨π³Maker 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. | 120.2From 41 benchmarks. Plenty of data behind it. | 106Β Β±15Coding IQ 105.6, 95% interval 90β121. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.20cache read $0.04 | $1.15 | 256K | 47.0Poor value (47.0). Costs more per task than its benchmark results predict. Plenty of data behind it. | 57 Β±9Chatty (57 Β±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. | βNot measured |
Step 3.7 Flash Inference by Novita AIπ¨π³Maker 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. | 120.2From 41 benchmarks. Plenty of data behind it. | 106Β Β±15Coding IQ 105.6, 95% interval 90β121. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.20cache read $0.04 | $1.15 | 262K | 47.0Poor value (47.0). Costs more per task than its benchmark results predict. Plenty of data behind it. | 57 Β±9Chatty (57 Β±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. | βNot measured |
Step 3.7 Flash Inference by StepFunπ¨π³Maker 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. | 120.2From 41 benchmarks. Plenty of data behind it. | 106Β Β±15Coding IQ 105.6, 95% interval 90β121. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.20cache read $0.04 | $1.15 | 256K | 47.0Poor value (47.0). Costs more per task than its benchmark results predict. Plenty of data behind it. | 57 Β±9Chatty (57 Β±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. | βNot measured |
Gemini 3 Pro Image πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | 119.8From 189 benchmarks. A fair amount of data behind it. | βNot enough data yet | $2.00cache read $0.20 | $12.00 | 66K | βNot enough data yet | βNot measured | βNot measured |
Gemini 3 Pro Image Inference by GoogleπΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | 119.8From 189 benchmarks. A fair amount of data behind it. | βNot enough data yet | $2.00cache read $0.20 | $12.00 | 131K | βNot enough data yet | βNot measured | βNot measured |
Gemini 3 Pro Image Inference by GoogleπΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | 119.8From 189 benchmarks. A fair amount of data behind it. | βNot enough data yet | $1.00cache read $0.10 | $6.00 | 66K | βNot enough data yet | βNot measured | βNot measured |
Ling 1t π¨π³Maker headquartered in CN, according to the company's own website. The citation is on the model page. | 119.8From 14 benchmarks. A fair amount of data behind it. | βNot enough data yet | $0.40 | $2.00 | βNot published by the provider | 58.0Good value (58.0). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
MiniMax M2.7 π¨π³Maker 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. | 119.8From 207 benchmarks. Plenty of data behind it. | 98Β Β±12Coding IQ 98.4, 95% interval 87β110. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.60cache read $0.054 | $1.80 | 197K | 50.7Fair value (50.7). Costs per task about what its benchmark results predict. Plenty of data behind it. | 61 Β±9Chatty (61 Β±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. | βNot measured |
MiniMax M2.7 Inference by Groqπ¨π³Maker 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. | 119.8From 207 benchmarks. Plenty of data behind it. | 98Β Β±12Coding IQ 98.4, 95% interval 87β110. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.60 | $1.80 | 197K | 50.7Fair value (50.7). Costs per task about what its benchmark results predict. Plenty of data behind it. | 61 Β±9Chatty (61 Β±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. | βNot measured |
Qwen3.6 35B A3B π¨π³Maker 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. | 119.8From 160 benchmarks. Plenty of data behind it. | 95Β Β±14Coding IQ 94.8, 95% interval 81β109. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.10cache read $0.025 | $0.90 | 262K | 38.5Poor value (38.5). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 144.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3.6 35B A3B Inference by Akashπ¨π³Maker 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. | 119.8From 160 benchmarks. Plenty of data behind it. | 95Β Β±14Coding IQ 94.8, 95% interval 81β109. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.10cache read $0.05 | $0.90 | 262K | 38.5Poor value (38.5). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 144.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3.6 35B A3B Inference by Darkbloomπ¨π³Maker 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. | 119.8From 160 benchmarks. Plenty of data behind it. | 95Β Β±14Coding IQ 94.8, 95% interval 81β109. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.05cache read $0.025 | $0.70 | 262K | 38.5Poor value (38.5). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 144.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3.6 35B A3B Inference by DeepInfraπ¨π³Maker 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. | 119.8From 160 benchmarks. Plenty of data behind it. | 95Β Β±14Coding IQ 94.8, 95% interval 81β109. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.10 | $0.95 | 262K | 38.5Poor value (38.5). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 144.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3.6 35B A3B Inference by DeepInfraπ¨π³Maker 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. | 119.8From 160 benchmarks. Plenty of data behind it. | 95Β Β±14Coding IQ 94.8, 95% interval 81β109. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.08 | $0.76 | 262K | 38.5Poor value (38.5). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 144.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3.6 35B A3B Inference by DeepInfraπ¨π³Maker 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. | 119.8From 160 benchmarks. Plenty of data behind it. | 95Β Β±14Coding IQ 94.8, 95% interval 81β109. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.15 | $1.425 | 262K | 38.5Poor value (38.5). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 144.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3.6 35B A3B Inference by Parasailπ¨π³Maker 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. | 119.8From 160 benchmarks. Plenty of data behind it. | 95Β Β±14Coding IQ 94.8, 95% interval 81β109. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.10cache read $0.025 | $0.90 | 262K | 38.5Poor value (38.5). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 144.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3.6 35B A3B Inference by Primeintellectπ¨π³Maker 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. | 119.8From 160 benchmarks. Plenty of data behind it. | 95Β Β±14Coding IQ 94.8, 95% interval 81β109. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.19cache read $0.17 | $1.70 | 262K | 38.5Poor value (38.5). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 144.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3.6 35B A3B Inference by SiliconFlowπ¨π³Maker 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. | 119.8From 160 benchmarks. Plenty of data behind it. | 95Β Β±14Coding IQ 94.8, 95% interval 81β109. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.24cache read $0.15 | $1.80 | 262K | 38.5Poor value (38.5). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 144.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3.6 35B A3B Inference by Veniceπ¨π³Maker 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. | 119.8From 160 benchmarks. Plenty of data behind it. | 95Β Β±14Coding IQ 94.8, 95% interval 81β109. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.10 | $1.00 | 256K | 38.5Poor value (38.5). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 144.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT 4 32k πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | β 120From 4 benchmarks. Little data behind it, so treat it as a rough estimate. | βNot enough data yet | $60.00 | $120.00 | βNot published by the provider | βNot enough data yet | βNot measured | βNot measured |
GPT-4.5 Preview πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 119.7From 206 benchmarks. Plenty of data behind it. | 88Β Β±11Coding IQ 88.3, 95% interval 77β100. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $75.00cache read $37.50 | $150.00 | βNot published by the provider | 23.1Poor value (23.1). Costs more per task than its benchmark results predict. A fair amount of data behind it. | βNot measured | 136.5Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
DeepSeek V3.1 π¨π³Maker 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. | 119.6From 122 benchmarks. Plenty of data behind it. | 94Β Β±8Coding IQ 94.3, 95% interval 86β103. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.00 | $0.00 | βNot published by the provider | 44.0Poor value (44.0). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 140.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
DeepSeek V3.1 π¨π³Maker 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. | 119.6From 122 benchmarks. Plenty of data behind it. | 94Β Β±8Coding IQ 94.3, 95% interval 86β103. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.25cache read $0.13 | $0.95 | 164K | 44.0Poor value (44.0). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 140.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
DeepSeek V3.1 Inference by Coreweaveπ¨π³Maker 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. | 119.6From 122 benchmarks. Plenty of data behind it. | 94Β Β±8Coding IQ 94.3, 95% interval 86β103. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.50cache read $0.20 | $1.50 | 161K | 44.0Poor value (44.0). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 140.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
DeepSeek V3.1 Inference by DeepInfraπ¨π³Maker 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. | 119.6From 122 benchmarks. Plenty of data behind it. | 94Β Β±8Coding IQ 94.3, 95% interval 86β103. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.25cache read $0.13 | $0.95 | 164K | 44.0Poor value (44.0). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 140.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
DeepSeek V3.1 Inference by SiliconFlowπ¨π³Maker 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. | 119.6From 122 benchmarks. Plenty of data behind it. | 94Β Β±8Coding IQ 94.3, 95% interval 86β103. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.27 | $1.00 | 164K | 44.0Poor value (44.0). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 140.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Mistral Medium 3.1 π«π·Maker headquartered in FR, according to the company's own website. The citation is on the model page. Newer version | 119.6From 89 benchmarks. Plenty of data behind it. | βNot enough data yet | $0.44cache read $0.044 | $2.20 | 131K | βNot enough data yet | βNot measured | βNot measured |
Mistral Medium 3.1 Inference by Mistral AIπ«π·Maker headquartered in FR, according to the company's own website. The citation is on the model page. Newer version | 119.6From 89 benchmarks. Plenty of data behind it. | βNot enough data yet | $0.44cache read $0.044 | $2.20 | 131K | βNot enough data yet | βNot measured | βNot measured |
Devstral 2 123b Instruct 2512 π«π·Maker headquartered in FR, 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. | 119.5From 25 benchmarks. Plenty of data behind it. | 97Β Β±9Coding IQ 97.1, 95% interval 88β106. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.40cache read $0.04 | $2.00 | 262K | 50.2Fair value (50.2). Costs per task about what its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
Devstral 2 123b Instruct 2512 Inference by Mistral AIπ«π·Maker headquartered in FR, 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. | 119.5From 25 benchmarks. Plenty of data behind it. | 97Β Β±9Coding IQ 97.1, 95% interval 88β106. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.40cache read $0.04 | $2.00 | 262K | 50.2Fair value (50.2). Costs per task about what its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
Gemini 2.5 Flash Lite Preview 09-2025 πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | 119.5From 21 benchmarks. A fair amount of data behind it. | βNot enough data yet | $0.10 | $0.40 | βNot published by the provider | 65.8Good value (65.8). Cheaper per task than its benchmark results predict. Plenty of data behind it. | βNot measured | βNot measured |
Kimi K2 0905 π¨π³Maker 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 | 119.4From 22 benchmarks. Plenty of data behind it. | β 93Β Β±11Coding IQ 92.5, 95% interval 82β103. 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. | $0.60 | $2.50 | 262K | 56.6Good value (56.6). Cheaper per task than its benchmark results predict. A fair amount of data behind it. | 41 Β±9Terse (41 Β±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. | βNot measured |
Kimi K2 0905 Inference by Novita AIπ¨π³Maker 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 | 119.4From 22 benchmarks. Plenty of data behind it. | β 93Β Β±11Coding IQ 92.5, 95% interval 82β103. 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. | $0.60 | $2.50 | 262K | 56.6Good value (56.6). Cheaper per task than its benchmark results predict. A fair amount of data behind it. | 41 Β±9Terse (41 Β±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. | βNot measured |
Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence)