Listings
All listings: Merge Gateway
Every listing for the models on page 6 of 8 of Merge Gateway, including each model's other versions and routes. Back to Merge Gateway
Sold by Merge Gateway
25 listings on this page sold by Merge Gateway.
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 |
|---|---|---|---|---|---|---|---|---|
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. | $0.56cache read $0.13 | $1.68 | 131K | 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 |
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 | 256K | 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 |
Qwen 3 235B A22B 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. | 119.3From 185 benchmarks. Plenty of data behind it. | 86Β Β±12Coding IQ 86.4, 95% interval 75β98. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.287cache read $0.0574 | $1.147 | 131K | 53.2Good value (53.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 143.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3 Coder 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. | 119.2From 84 benchmarks. Plenty of data behind it. | 93Β Β±6Coding IQ 92.6, 95% interval 87β98. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.22 | $1.80 | 131K | 34.9Poor value (34.9). Costs more per task than its benchmark results predict. Plenty of data behind it. | 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 |
MiniMax M2 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. | 118.9From 111 benchmarks. Plenty of data behind it. | 95Β Β±7Coding IQ 95.4, 95% interval 88β103. 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 | $1.20cache write $0.375 | 205K | 57.8Good value (57.8). Cheaper per task than its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
Seed 2.0 Minibytedance/dola-seed-2.0-minibytedance/dola-seed-2.0-mini | 118.6From 76 benchmarks. Plenty of data behind it. | βNot enough data yet | $0.12 | $0.50 | 256K | 45.0Poor value (45.0). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
GPT OSS 120B 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. | 118.2From 424 benchmarks. Plenty of data behind it. | 86Β Β±5Coding IQ 85.5, 95% interval 80β91. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.05 | $0.25 | 128K | 58.4Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 37 Β±11Terse (37 Β±11). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 140.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3 Coder Next 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. | 118.2From 27 benchmarks. Plenty of data behind it. | 98Β Β±7Coding IQ 97.7, 95% interval 90β105. 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 $0.075 | $0.80 | 262K | 36.8Poor value (36.8). Costs more per task than its benchmark results predict. Plenty of data behind it. | 54 Β±9Typical (54 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 |
Mistral Large 3 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. | 118.1From 60 benchmarks. Plenty of data behind it. | 86Β Β±8Coding IQ 86.4, 95% interval 79β94. 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 $0.05 | $1.50 | 256K | 38.5Poor value (38.5). Costs more per task than 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. | βNot measured |
Qwen3 Next 80B A3B 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. | 118.1From 154 benchmarks. Plenty of data behind it. | 92Β Β±11Coding IQ 91.7, 95% interval 81β102. A fair amount 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 | $1.10 | 131K | 47.6Fair value (47.6). Costs per task about what its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
GLM 4.5 Air 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. | 118.0From 210 benchmarks. Plenty of data behind it. | 93Β Β±7Coding IQ 93.2, 95% interval 86β100. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.20cache read $0.03 | $1.10 | 128K | 42.6Poor value (42.6). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | βNot measured |
Command A 03 2025 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. | β 118From 21 benchmarks. Little data behind it, so treat it as a rough estimate. | βNot enough data yet | $2.50 | $10.00 | 256K | 20.8Poor value (20.8). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | βNot measured |
Devstral Medium 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. | β 117From 10 benchmarks. Little data behind it, so treat it as a rough estimate. | βNot enough data yet | $0.40 | $2.00 | 262K | 42.0Poor value (42.0). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
Codestral 25.01 π«π·Maker headquartered in FR, according to the company's own website. The citation is on the model page. | 116.6From 10 benchmarks. A fair amount of data behind it. | βNot enough data yet | $0.30 | $0.90 | 256K | 37.8Poor value (37.8). Costs more per task than its benchmark results predict. A fair amount 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. | βNot measured |
Qwen Plus π¨π³Maker headquartered in CN, according to the company's own website. The citation is on the model page. | 116.4From 42 benchmarks. Plenty of data behind it. | βNot enough data yet | $0.115cache read $0.023 | $0.287cache write $0.1438 | 1M | 43.6Poor value (43.6). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | βNot measured |
GPT-4 Turbo πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | 116.2From 380 benchmarks. Plenty of data behind it. | 83Β Β±10Coding IQ 83.0, 95% interval 73β93. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $10.00 | $30.00 | 128K | 34.4Poor value (34.4). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | βNot measured | 125.8Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GLM 4.7 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. Newer version | 116.1From 186 benchmarks. Plenty of data behind it. | 83Β Β±10Coding IQ 82.8, 95% interval 73β93. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.07 | $0.40 | 200K | 44.1Poor value (44.1). Costs more per task than its benchmark results predict. A fair amount of data behind it. | βNot measured | βNot measured |
GLM 4.5V 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. | 115.8From 62 benchmarks. A fair amount of data behind it. | βNot enough data yet | $0.60cache read $0.11 | $1.80 | 128K | 36.7Poor value (36.7). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | βNot measured |
Gemini 3.1 Flash Image πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. | 115.7From 40 benchmarks. A fair amount of data behind it. | βNot enough data yet | $0.50 | $3.00 | 33K | βNot enough data yet | βNot measured | βNot measured |
Qwen3 32B 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. | 115.6From 234 benchmarks. Plenty of data behind it. | 85Β Β±9Coding IQ 84.8, 95% interval 76β93. 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 | $0.60 | 131K | 35.8Poor value (35.8). Costs more per task than its benchmark results predict. A fair amount of data behind it. | βNot measured | 136.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-4.1 mini πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 115.4From 375 benchmarks. Plenty of data behind it. | 91Β Β±8Coding IQ 91.3, 95% interval 84β99. 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 $0.10 | $1.60 | 1M | 46.2Poor value (46.2). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | 134.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
GPT-5 nano πΊπΈMaker headquartered in US, according to the company's own website. The citation is on the model page. Newer version | 115.3From 332 benchmarks. Plenty of data behind it. | 93Β Β±13Coding IQ 92.7, 95% interval 80β106. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.05cache read $0.005 | $0.40 | 400K | 46.3Poor value (46.3). Costs more per task than its benchmark results predict. Plenty of data behind it. | 51 Β±9Typical (51 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. | 139.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Mistral Medium 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. | 114.7From 48 benchmarks. Plenty of data behind it. | β 91Β Β±14Coding IQ 90.8, 95% interval 76β105. 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. | $0.40 | $2.00 | 262K | 24.3Poor value (24.3). Costs more per task than its benchmark results predict. A fair amount 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. | βNot measured |
DeepSeek V3 π¨π³Maker headquartered in CN, according to the company's own website. The citation is on the model page. Newer version | 114.5From 492 benchmarks. Plenty of data behind it. | 85Β Β±7Coding IQ 85.3, 95% interval 79β92. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.27 | $1.10 | 164K | 43.6Poor value (43.6). Costs more per task than 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. | 135.3Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Nemotron 3.5 Lightning 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. | 114.4From 121 benchmarks. Plenty of data behind it. | 92Β Β±22Coding IQ 91.7, 95% interval 70β114. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | β | β | 1M | 30.6Poor value (30.6). Costs more per task than its benchmark results predict. Plenty of data behind it. | βNot measured | βNot measured |
Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence)