Listings
All listings: OpenRouter
Every listing for the models on page 12 of 26 of OpenRouter, including each model's other versions and routes. Back to OpenRouter
Sold by OpenRouter
58 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 |
|---|---|---|---|---|---|---|---|---|
Qwen Max π¨π³Maker headquartered in CN, according to the company's own website. The citation is on the model page. | 111.6From 56 benchmarks. Plenty of data behind it. | β 82Β Β±14Coding IQ 81.7, 95% interval 68β95. 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. | $1.60cache read $0.64 | $6.40 | βNot published by the provider | 42.6Poor value (42.6). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
qwen3-vl-30b 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. | 111.5From 81 benchmarks. Plenty of data behind it. | βNot enough data yet | $0.13 | $1.56 | βNot published by the provider | 45.6Poor value (45.6). Costs more per task than its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
qwen3-vl-30b 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. | 111.5From 81 benchmarks. Plenty of data behind it. | βNot enough data yet | $0.15 | $0.60 | 262K | 45.6Poor value (45.6). Costs more per task than its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
qwen3-vl-30b 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. | 111.5From 81 benchmarks. Plenty of data behind it. | βNot enough data yet | $0.15 | $0.60 | 262K | 45.6Poor value (45.6). Costs more per task than its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
qwen3-vl-30b 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. | 111.5From 81 benchmarks. Plenty of data behind it. | βNot enough data yet | $0.20 | $0.70 | 131K | 45.6Poor value (45.6). Costs more per task than its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
qwen3-vl-30b 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. | 111.5From 81 benchmarks. Plenty of data behind it. | βNot enough data yet | $0.29 | $1.00 | 262K | 45.6Poor value (45.6). Costs more per task than its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
GPT OSS 20b πΊπΈMaker 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. | 111.3From 251 benchmarks. Plenty of data behind it. | 88Β Β±12Coding IQ 88.2, 95% interval 76β101. Plenty 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 | 48.9Fair value (48.9). Costs per task about what its benchmark results predict. Plenty of data behind it. | 40 Β±9Terse (40 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. | 134.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT OSS 20b πΊπΈMaker 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. | 111.3From 251 benchmarks. Plenty of data behind it. | 88Β Β±12Coding IQ 88.2, 95% interval 76β101. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.017cache read $0.0085 | $0.085 | 131K | 48.9Fair value (48.9). Costs per task about what its benchmark results predict. Plenty of data behind it. | 40 Β±9Terse (40 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. | 134.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT OSS 20b πΊπΈMaker 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. | 111.3From 251 benchmarks. Plenty of data behind it. | 88Β Β±12Coding IQ 88.2, 95% interval 76β101. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.024 | $0.112 | 131K | 48.9Fair value (48.9). Costs per task about what its benchmark results predict. Plenty of data behind it. | 40 Β±9Terse (40 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. | 134.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT OSS 20b Inference by AkashπΊπΈMaker 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. | 111.3From 251 benchmarks. Plenty of data behind it. | 88Β Β±12Coding IQ 88.2, 95% interval 76β101. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.018cache read $0.009 | $0.09 | 131K | 48.9Fair value (48.9). Costs per task about what its benchmark results predict. Plenty of data behind it. | 40 Β±9Terse (40 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. | 134.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT OSS 20b Inference by Amazon Web ServicesπΊπΈMaker 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. | 111.3From 251 benchmarks. Plenty of data behind it. | 88Β Β±12Coding IQ 88.2, 95% interval 76β101. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.07 | $0.15 | 131K | 48.9Fair value (48.9). Costs per task about what its benchmark results predict. Plenty of data behind it. | 40 Β±9Terse (40 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. | 134.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT OSS 20b Inference by Amazon Web ServicesπΊπΈMaker 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. | 111.3From 251 benchmarks. Plenty of data behind it. | 88Β Β±12Coding IQ 88.2, 95% interval 76β101. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.07 | $0.15 | 131K | 48.9Fair value (48.9). Costs per task about what its benchmark results predict. Plenty of data behind it. | 40 Β±9Terse (40 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. | 134.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT OSS 20b Inference by CoreweaveπΊπΈMaker 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. | 111.3From 251 benchmarks. Plenty of data behind it. | 88Β Β±12Coding IQ 88.2, 95% interval 76β101. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.03 | $0.13 | 131K | 48.9Fair value (48.9). Costs per task about what its benchmark results predict. Plenty of data behind it. | 40 Β±9Terse (40 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. | 134.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT OSS 20b Inference by DarkbloomπΊπΈMaker 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. | 111.3From 251 benchmarks. Plenty of data behind it. | 88Β Β±12Coding IQ 88.2, 95% interval 76β101. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.017cache read $0.0085 | $0.085 | 131K | 48.9Fair value (48.9). Costs per task about what its benchmark results predict. Plenty of data behind it. | 40 Β±9Terse (40 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. | 134.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT OSS 20b Inference by DeepInfraπΊπΈMaker 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. | 111.3From 251 benchmarks. Plenty of data behind it. | 88Β Β±12Coding IQ 88.2, 95% interval 76β101. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.024 | $0.112 | 131K | 48.9Fair value (48.9). Costs per task about what its benchmark results predict. Plenty of data behind it. | 40 Β±9Terse (40 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. | 134.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT OSS 20b Inference by DeepInfraπΊπΈMaker 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. | 111.3From 251 benchmarks. Plenty of data behind it. | 88Β Β±12Coding IQ 88.2, 95% interval 76β101. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.03cache read $0.015 | $0.14 | 131K | 48.9Fair value (48.9). Costs per task about what its benchmark results predict. Plenty of data behind it. | 40 Β±9Terse (40 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. | 134.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT OSS 20b Inference by DeepInfraπΊπΈMaker 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. | 111.3From 251 benchmarks. Plenty of data behind it. | 88Β Β±12Coding IQ 88.2, 95% interval 76β101. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.024cache read $0.012 | $0.112 | 131K | 48.9Fair value (48.9). Costs per task about what its benchmark results predict. Plenty of data behind it. | 40 Β±9Terse (40 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. | 134.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT OSS 20b Inference by GroqπΊπΈMaker 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. | 111.3From 251 benchmarks. Plenty of data behind it. | 88Β Β±12Coding IQ 88.2, 95% interval 76β101. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.075cache read $0.0375 | $0.30 | 131K | 48.9Fair value (48.9). Costs per task about what its benchmark results predict. Plenty of data behind it. | 40 Β±9Terse (40 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. | 134.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT OSS 20b Inference by ParasailπΊπΈMaker 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. | 111.3From 251 benchmarks. Plenty of data behind it. | 88Β Β±12Coding IQ 88.2, 95% interval 76β101. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.03cache read $0.02 | $0.15 | 131K | 48.9Fair value (48.9). Costs per task about what its benchmark results predict. Plenty of data behind it. | 40 Β±9Terse (40 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. | 134.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT OSS 20b Inference by PrimeintellectπΊπΈMaker 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. | 111.3From 251 benchmarks. Plenty of data behind it. | 88Β Β±12Coding IQ 88.2, 95% interval 76β101. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.20cache read $0.18 | $0.30 | 131K | 48.9Fair value (48.9). Costs per task about what its benchmark results predict. Plenty of data behind it. | 40 Β±9Terse (40 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. | 134.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT OSS 20b Inference by SiliconFlowπΊπΈMaker 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. | 111.3From 251 benchmarks. Plenty of data behind it. | 88Β Β±12Coding IQ 88.2, 95% interval 76β101. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.04 | $0.18 | 131K | 48.9Fair value (48.9). Costs per task about what its benchmark results predict. Plenty of data behind it. | 40 Β±9Terse (40 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. | 134.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen 2.5 VL 32B Instruct 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 | 111.3From 25 benchmarks. A fair amount of data behind it. | βNot enough data yet | β | β | βNot published by the provider | 47.9Fair value (47.9). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Qwen 2.5 VL 32B Instruct 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 | 111.3From 25 benchmarks. A fair amount of data behind it. | βNot enough data yet | $0.00 | $0.00 | βNot published by the provider | 47.9Fair value (47.9). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Claude 3 Opus πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 111.1From 307 benchmarks. Plenty of data behind it. | 80Β Β±10Coding IQ 79.6, 95% interval 70β89. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $15.00cache read $1.50 | $75.00cache write $18.75 | βNot published by the provider | 38.0Poor value (38.0). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | 125.6Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Granite 4.1 8B πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | β 111From 36 benchmarks. Little data behind it, so treat it as a rough estimate. | βNot enough data yet | $0.05 | $0.10 | βNot published by the provider | 46.9Poor value (46.9). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Granite 4.2 8B πΊπΈMaker 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. | 110.9From 57 benchmarks. Plenty of data behind it. | 90Β Β±17Coding IQ 90.2, 95% interval 73β108. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.06cache read $0.015 | $0.15 | 131K | βNot enough data yet | βNot measured | βNot measured |
Granite 4.2 8B Inference by CoreweaveπΊπΈMaker 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. | 110.9From 57 benchmarks. Plenty of data behind it. | 90Β Β±17Coding IQ 90.2, 95% interval 73β108. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.06cache read $0.015 | $0.15 | 131K | βNot enough data yet | βNot measured | βNot measured |
Granite 4.2 8B Inference by DeepInfraπΊπΈMaker 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. | 110.9From 57 benchmarks. Plenty of data behind it. | 90Β Β±17Coding IQ 90.2, 95% interval 73β108. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.06cache read $0.015 | $0.25 | 131K | βNot enough data yet | βNot measured | βNot measured |
Nemotron Nano 9b V2 πΊπΈMaker 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. | 110.6From 19 benchmarks. Plenty of data behind it. | βNot enough data yet | $0.04 | $0.16 | βNot published by the provider | 52.1Fair value (52.1). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Nemotron Nano 9b V2 πΊπΈMaker 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. | 110.6From 19 benchmarks. Plenty of data behind it. | βNot enough data yet | $0.00 | $0.00 | βNot published by the provider | 52.1Fair value (52.1). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Hunyuan A13B Instruct 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 111From 9 benchmarks. Little data behind it, so treat it as a rough estimate. | βNot enough data yet | β | β | βNot published by the provider | 47.0Poor value (47.0). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Hunyuan A13B Instruct 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 111From 9 benchmarks. Little data behind it, so treat it as a rough estimate. | βNot enough data yet | $0.00 | $0.00 | βNot published by the provider | 47.0Poor value (47.0). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Claude Haiku 3.5 πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 110.1From 238 benchmarks. Plenty of data behind it. | 79Β Β±8Coding IQ 79.4, 95% interval 72β87. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.80 | $4.00 | βNot published by the provider | 19.4Poor value (19.4). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 124.8Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Claude Haiku 3.5 πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 110.1From 238 benchmarks. Plenty of data behind it. | 79Β Β±8Coding IQ 79.4, 95% interval 72β87. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.80 | $4.00 | βNot published by the provider | 19.4Poor value (19.4). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 124.8Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
DeepSeek R1 0528 Qwen3 8B 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 110From 9 benchmarks. Little data behind it, so treat it as a rough estimate. | βNot enough data yet | $0.05 | $0.10 | βNot published by the provider | 43.4Poor value (43.4). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
DeepSeek R1 0528 Qwen3 8B 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 110From 9 benchmarks. Little data behind it, so treat it as a rough estimate. | βNot enough data yet | $0.00 | $0.00 | βNot published by the provider | 43.4Poor value (43.4). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
β 110From 21 benchmarks. Little data behind it, so treat it as a rough estimate. | βNot enough data yet | $0.25 | $1.00 | βNot published by the provider | 51.0Fair value (51.0). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured | |
109.8From 29 benchmarks. Plenty of data behind it. | βNot enough data yet | $0.07 | $0.26 | βNot published by the provider | 53.1Good value (53.1). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured | |
Llama 3.1 Nemotron 70B Instruct πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | 109.8From 42 benchmarks. Plenty of data behind it. | βNot enough data yet | $0.12 | $0.30 | βNot published by the provider | 57.1Good value (57.1). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
MiniMax 01 π¨π³Maker headquartered in CN, according to the company's own website. The citation is on the model page. | β 110From 8 benchmarks. Little data behind it, so treat it as a rough estimate. | βNot enough data yet | $0.20 | $1.10 | 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 |
Llama 3.1 405B πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | 109.7From 26 benchmarks. Plenty of data behind it. | βNot enough data yet | $2.00 | $2.00 | βNot published by the provider | βNot enough data yet | βNot measured | 127.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Nemotron 3 Nano 30B A3B πΊπΈMaker 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. | 109.7From 137 benchmarks. Plenty of data behind it. | 91Β Β±14Coding IQ 91.4, 95% interval 77β106. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.06cache read $0.012 | $0.24 | 262K | 36.7Poor value (36.7). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Nemotron 3 Nano 30B A3B Inference by Nebius AI StudioπΊπΈMaker 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. | 109.7From 137 benchmarks. Plenty of data behind it. | 91Β Β±14Coding IQ 91.4, 95% interval 77β106. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.06cache read $0.012 | $0.24 | 262K | 36.7Poor value (36.7). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Gemini 1.5 Flash πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 109.6From 160 benchmarks. Plenty of data behind it. | β 77Β Β±12Coding IQ 76.5, 95% interval 65β88. 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. | $0.075cache read $0.0188 | $0.30cache write $0.1583 | βNot published by the provider | βNot enough data yet | βNot measured | 127Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Phi 4 Reasoning Plus πΊπΈMaker 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. | 109.5From 16 benchmarks. A fair amount of data behind it. | βNot enough data yet | $0.07 | $0.35 | βNot published by the provider | 42.3Poor value (42.3). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Magistral Small 2509 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. | 109.2From 42 benchmarks. Plenty of data behind it. | βNot enough data yet | $0.50 | $1.50 | βNot published by the provider | 27.6Poor value (27.6). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | βNot measured |
Gemini 3.1 Flash-Lite Image πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | β 109From 14 benchmarks. Little data behind it, so treat it as a rough estimate. | βNot enough data yet | $0.25 | $1.50 | 66K | 37.6Poor value (37.6). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Gemini 3.1 Flash-Lite Image Inference by GoogleπΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | β 109From 14 benchmarks. Little data behind it, so treat it as a rough estimate. | βNot enough data yet | $0.25 | $1.50 | 66K | 37.6Poor value (37.6). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Gemini 3.1 Flash-Lite Image Inference by GoogleπΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | β 109From 14 benchmarks. Little data behind it, so treat it as a rough estimate. | βNot enough data yet | $0.25 | $1.50 | 66K | 37.6Poor value (37.6). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Qwen Turbo π¨π³Maker headquartered in CN, according to the company's own website. The citation is on the model page. | 109.0From 5 benchmarks. A fair amount of data behind it. | βNot enough data yet | $0.05cache read $0.02 | $0.20 | βNot published by the provider | 61.6Good value (61.6). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Qwen: Qwen3 VL 8B Thinking π¨π³Maker headquartered in CN, according to the company's own website. The citation is on the model page. | 109.0From 103 benchmarks. Plenty of data behind it. | βNot enough data yet | $0.117 | $1.365 | βNot published by the provider | 41.1Poor value (41.1). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Qwen2.5 72B Instruct π¨π³Maker headquartered in CN, according to the company's own website. The citation is on the model page. | 108.9From 216 benchmarks. Plenty of data behind it. | 80Β Β±22Coding IQ 79.9, 95% interval 58β102. 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 | 35.0Poor value (35.0). Costs more per task than its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
Qwen2.5 72B Instruct π¨π³Maker headquartered in CN, according to the company's own website. The citation is on the model page. | 108.9From 216 benchmarks. Plenty of data behind it. | 80Β Β±22Coding IQ 79.9, 95% interval 58β102. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.36 | $0.40 | 33K | 35.0Poor value (35.0). Costs more per task than its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
Qwen2.5 72B Instruct Inference by DeepInfraπ¨π³Maker headquartered in CN, according to the company's own website. The citation is on the model page. | 108.9From 216 benchmarks. Plenty of data behind it. | 80Β Β±22Coding IQ 79.9, 95% interval 58β102. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.36 | $0.40 | 33K | 35.0Poor value (35.0). Costs more per task than its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
Qwen2.5 72B Instruct Inference by DeepInfraπ¨π³Maker headquartered in CN, according to the company's own website. The citation is on the model page. | 108.9From 216 benchmarks. Plenty of data behind it. | 80Β Β±22Coding IQ 79.9, 95% interval 58β102. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.288 | $0.32 | 33K | 35.0Poor value (35.0). Costs more per task than its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
Qwen2.5 VL 72B Instruct 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. | 108.8From 97 benchmarks. Plenty of data behind it. | βNot enough data yet | $0.00 | $0.00 | βNot published by the provider | 41.2Poor value (41.2). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Qwen2.5 VL 72B Instruct 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. | 108.8From 97 benchmarks. Plenty of data behind it. | βNot enough data yet | $0.80cache read $0.40 | $1.00 | 128K | 41.2Poor value (41.2). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Qwen2.5 VL 72B Instruct 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. | 108.8From 97 benchmarks. Plenty of data behind it. | βNot enough data yet | $0.80cache read $0.40 | $1.00 | 128K | 41.2Poor value (41.2). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence)