Listings
All listings: Novita AI
Every listing for the models on page 4 of 8 of Novita AI, including each model's other versions and routes. Back to Novita AI
Sold by Novita AI
26 listings on this page sold by Novita AI.
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 |
|---|---|---|---|---|---|---|---|---|
Kimi K2 0905 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. Newer version | 119.4From 22 benchmarks. Plenty of data behind it. | ≈ 93 ±11Coding IQ 92.5, 95% interval 82–103. 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 ±9Terse (41 ±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. | —Not measured |
Qwen 3 235B A22B 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 119.3From 185 benchmarks. Plenty of data behind it. | 86 ±12Coding IQ 86.4, 95% interval 75–98. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.20 | $0.80 | 41K | 53.2Good value (53.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | —Not measured | 143.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
MiMo V2 Flash 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. Newer version | 119.2From 153 benchmarks. Plenty of data behind it. | 95 ±11Coding IQ 94.7, 95% interval 84–106. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.11cache read $0.024 | $0.33 | 262K | 53.7Good value (53.7). Cheaper per task than its benchmark results predict. A fair amount of data behind it. | 53 ±9Typical (53 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | —Not measured |
Qwen3 Coder 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 119.2From 84 benchmarks. Plenty of data behind it. | 93 ±6Coding IQ 92.6, 95% interval 87–98. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.38 | $1.55 | 262K | 34.9Poor value (34.9). Costs more per task than its benchmark results predict. Plenty of data behind it. | 39 ±9Terse (39 ±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. | —Not measured |
MiniMax M2 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 118.9From 111 benchmarks. Plenty of data behind it. | 95 ±7Coding IQ 95.4, 95% interval 88–103. 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 | $1.20 | 205K | 57.8Good value (57.8). Cheaper per task than its benchmark results predict. A fair amount of data behind it. | —Not measured | —Not measured |
GLM 4.6v 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 118.4From 45 benchmarks. A fair amount of data behind it. | —Not enough data yet | $0.30cache read $0.055 | $0.90 | 131K | 44.9Poor value (44.9). Costs more per task than its benchmark results predict. A fair amount of data behind it. | —Not measured | —Not measured |
GPT OSS 120B 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 118.2From 424 benchmarks. Plenty of data behind it. | 86 ±5Coding IQ 85.5, 95% interval 80–91. 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 | 131K | 58.4Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 37 ±11Terse (37 ±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 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 118.2From 27 benchmarks. Plenty of data behind it. | 98 ±7Coding IQ 97.7, 95% interval 90–105. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.20 | $1.50 | 262K | 36.8Poor value (36.8). Costs more per task than its benchmark results predict. Plenty of data behind it. | 54 ±9Typical (54 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | —Not measured |
Qwen3 Next 80B A3B Instruct 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 118.1From 154 benchmarks. Plenty of data behind it. | 92 ±11Coding IQ 91.7, 95% interval 81–102. 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 | $1.50 | 131K | 47.6Fair value (47.6). Costs per task about what its benchmark results predict. A fair amount of data behind it. | —Not measured | —Not measured |
GLM 4.5 Air 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 118.0From 210 benchmarks. Plenty of data behind it. | 93 ±7Coding IQ 93.2, 95% interval 86–100. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.13cache read $0.025 | $0.85 | 131K | 42.6Poor value (42.6). Costs more per task than its benchmark results predict. Plenty of data behind it. | —Not measured | —Not measured |
Qwen3 Coder 30B A3B Instruct 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 116.2From 18 benchmarks. A fair amount of data behind it. | —Not enough data yet | $0.07 | $0.27 | 160K | 54.3Good value (54.3). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | —Not measured |
GLM 4.7 Flash 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. Newer version | 116.1From 186 benchmarks. Plenty of data behind it. | 83 ±10Coding IQ 82.8, 95% interval 73–93. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.07cache read $0.01 | $0.40 | 200K | 44.1Poor value (44.1). Costs more per task than its benchmark results predict. A fair amount of data behind it. | —Not measured | —Not measured |
GLM 4.5V 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 115.8From 62 benchmarks. A fair amount of data behind it. | —Not enough data yet | $0.60cache read $0.11 | $1.80 | 66K | 36.7Poor value (36.7). Costs more per task than its benchmark results predict. Plenty of data behind it. | —Not measured | —Not measured |
Qwen3 32B 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 115.6From 234 benchmarks. Plenty of data behind it. | 85 ±9Coding IQ 84.8, 95% interval 76–93. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.10 | $0.45 | 41K | 35.8Poor value (35.8). Costs more per task than its benchmark results predict. A fair amount of data behind it. | —Not measured | 136.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
DeepSeek V3 🇨🇳Maker 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. | 85 ±7Coding IQ 85.3, 95% interval 79–92. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.89 | $0.89 | 64K | 43.6Poor value (43.6). Costs more per task than its benchmark results predict. Plenty of data behind it. | 47 ±9Typical (47 ±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 |
DeepSeek V3 🇨🇳Maker 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. | 85 ±7Coding IQ 85.3, 95% interval 79–92. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.27cache read $0.135 | $1.12 | 164K | 43.6Poor value (43.6). Costs more per task than its benchmark results predict. Plenty of data behind it. | 47 ±9Typical (47 ±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 |
Qwen3 30B A3B 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 114.4From 142 benchmarks. Plenty of data behind it. | 90 ±10Coding IQ 90.3, 95% interval 80–101. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.09 | $0.45 | 41K | 35.8Poor value (35.8). Costs more per task than its benchmark results predict. Plenty of data behind it. | —Not measured | 138.3Epoch Capabilities IndexSetting: thinkingSource: Epoch AI Benchmarking Hub API |
GLM 4 32B 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. | 113.3From 10 benchmarks. A fair amount of data behind it. | —Not enough data yet | $0.55 | $1.66 | 32K | 58.4Good value (58.4). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | —Not measured |
Qwen3 8B 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 112.4From 182 benchmarks. Plenty of data behind it. | 80 ±12Coding IQ 80.3, 95% interval 68–92. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.035 | $0.138 | 128K | 58.0Good value (58.0). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | 132.1Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Llama 4 Maverick 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 111.8From 350 benchmarks. Plenty of data behind it. | 78 ±9Coding IQ 78.1, 95% interval 69–87. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.27 | $0.85 | 1M | 42.1Poor value (42.1). Costs more per task than its benchmark results predict. Plenty of data behind it. | 38 ±9Terse (38 ±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. | 132.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
qwen3-vl-30b 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 111.5From 81 benchmarks. Plenty of data behind it. | —Not enough data yet | $0.20 | $1.00 | 131K | 45.6Poor value (45.6). Costs more per task than its benchmark results predict. A fair amount of data behind it. | —Not measured | —Not measured |
qwen3-vl-30b 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 111.5From 81 benchmarks. Plenty of data behind it. | —Not 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. | —Not measured | —Not measured |
GPT OSS 20b 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 111.3From 251 benchmarks. Plenty of data behind it. | 88 ±12Coding IQ 88.2, 95% interval 76–101. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.04 | $0.15 | 131K | 48.9Fair value (48.9). Costs per task about what its benchmark results predict. Plenty of data behind it. | 40 ±9Terse (40 ±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 134.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Qwen3 Embedding 0.6B 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 110.5From 192 benchmarks. Plenty of data behind it. | —Not enough data yet | $0.07 | — | 33K | —Not enough data yet | —Not measured | —Not measured |
DeepSeek R1 0528 Qwen3 8B 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | ≈ 110From 9 benchmarks. Little data behind it, so treat it as a rough estimate. | —Not enough data yet | $0.06 | $0.09 | 128K | 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. | —Not measured | —Not measured |
Nemotron 3 Nano 30B A3B 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 109.7From 137 benchmarks. Plenty of data behind it. | 91 ±14Coding IQ 91.4, 95% interval 77–106. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.05 | $0.20 | 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. | —Not measured | —Not measured |
Run by Novita AI
36 listings on this page where Novita AI runs the model. Another provider sells these listings.
Select a score for details.
About these columns
- llminfo IQ
- llminfo IQ (capability index). How llminfo IQ is computed
- Coding IQ
- llminfo Coding IQ (coding capability, its own scale). How Coding IQ is computed
- Value
- Value for money. 50 means as expected for its results. Above 50, cheaper per task. How Value for money works
- Yap
- Tokens used per coding task vs. models with the same results. Includes agents that write code. Chattiness on maths doesn't predict it. How the Yap score works
- Epoch Capabilities Index
- Epoch Capabilities Index benchmark details
| Model | Provider | llminfo IQ | Coding IQ | Input $/1M | Output $/1M | Max Input Tokens | Value | Yap | Epoch Capabilities Index |
|---|---|---|---|---|---|---|---|---|---|
Kimi K2 0905 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. Newer version | 119.4From 22 benchmarks. Plenty of data behind it. | ≈ 93 ±11Coding IQ 92.5, 95% interval 82–103. 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 ±9Terse (41 ±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. | —Not measured | |
Qwen 3 235B A22B 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 119.3From 185 benchmarks. Plenty of data behind it. | 86 ±12Coding IQ 86.4, 95% interval 75–98. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.20 | $0.80 | 41K | 53.2Good value (53.2). Cheaper per task than its benchmark results predict. Plenty of data behind it. | —Not measured | 143.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
MiMo V2 Flash 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. Newer version | 119.2From 153 benchmarks. Plenty of data behind it. | 95 ±11Coding IQ 94.7, 95% interval 84–106. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.10 | $0.30 | 262K | 53.7Good value (53.7). Cheaper per task than its benchmark results predict. A fair amount of data behind it. | 53 ±9Typical (53 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | —Not measured | |
Qwen3 Coder 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 119.2From 84 benchmarks. Plenty of data behind it. | 93 ±6Coding IQ 92.6, 95% interval 87–98. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.30 | $1.30 | 262K | 34.9Poor value (34.9). Costs more per task than its benchmark results predict. Plenty of data behind it. | 39 ±9Terse (39 ±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. | —Not measured | |
MiniMax M2 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 118.9From 111 benchmarks. Plenty of data behind it. | 95 ±7Coding IQ 95.4, 95% interval 88–103. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.30 | $1.20 | 205K | 57.8Good value (57.8). Cheaper per task than its benchmark results predict. A fair amount of data behind it. | —Not measured | —Not measured | |
GLM 4.6v 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 118.4From 45 benchmarks. A fair amount of data behind it. | —Not enough data yet | $0.30cache read $0.055 | $0.90 | —Not published by the provider | 44.9Poor value (44.9). Costs more per task than its benchmark results predict. A fair amount of data behind it. | —Not measured | —Not measured | |
GLM 4.6v 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 118.4From 45 benchmarks. A fair amount of data behind it. | —Not enough data yet | $0.30cache read $0.055 | $0.90 | 131K | 44.9Poor value (44.9). Costs more per task than its benchmark results predict. A fair amount of data behind it. | —Not measured | —Not measured | |
GPT OSS 120B 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 118.2From 424 benchmarks. Plenty of data behind it. | 86 ±5Coding IQ 85.5, 95% interval 80–91. 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 | 131K | 58.4Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it. | 37 ±11Terse (37 ±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 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 118.2From 27 benchmarks. Plenty of data behind it. | 98 ±7Coding IQ 97.7, 95% interval 90–105. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.20 | $1.50 | 262K | 36.8Poor value (36.8). Costs more per task than its benchmark results predict. Plenty of data behind it. | 54 ±9Typical (54 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | —Not measured | |
Qwen3 Next 80B A3B Instruct 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 118.1From 154 benchmarks. Plenty of data behind it. | 92 ±11Coding IQ 91.7, 95% interval 81–102. 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 | $1.50 | 131K | 47.6Fair value (47.6). Costs per task about what its benchmark results predict. A fair amount of data behind it. | —Not measured | —Not measured | |
GLM 4.5 Air 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 118.0From 210 benchmarks. Plenty of data behind it. | 93 ±7Coding IQ 93.2, 95% interval 86–100. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.13 | $0.85 | 131K | 42.6Poor value (42.6). Costs more per task than its benchmark results predict. Plenty of data behind it. | —Not measured | —Not measured | |
GLM 4.5 Air 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 118.0From 210 benchmarks. Plenty of data behind it. | 93 ±7Coding IQ 93.2, 95% interval 86–100. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.1365cache read $0.0263 | $0.8925 | —Not published by the provider | 42.6Poor value (42.6). Costs more per task than its benchmark results predict. Plenty of data behind it. | —Not measured | —Not measured | |
GLM 4.5 Air 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 118.0From 210 benchmarks. Plenty of data behind it. | 93 ±7Coding IQ 93.2, 95% interval 86–100. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.1365cache read $0.0263 | $0.8925 | —Not published by the provider | 42.6Poor value (42.6). Costs more per task than its benchmark results predict. Plenty of data behind it. | —Not measured | —Not measured | |
GLM 4.5 Air 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 118.0From 210 benchmarks. Plenty of data behind it. | 93 ±7Coding IQ 93.2, 95% interval 86–100. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.13cache read $0.025 | $0.85 | 131K | 42.6Poor value (42.6). Costs more per task than its benchmark results predict. Plenty of data behind it. | —Not measured | —Not measured | |
Kimi K2 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. Newer version | 117.4From 30 benchmarks. Plenty of data behind it. | ≈ 95 ±16staleCoding IQ 95.1, 95% interval 80–111. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ. Its newest result is over a year old. | $0.57 | $2.30 | —Not published by the provider | 52.3Fair value (52.3). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | 141.3Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
Kimi K2 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. Newer version | 117.4From 30 benchmarks. Plenty of data behind it. | ≈ 95 ±16staleCoding IQ 95.1, 95% interval 80–111. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ. Its newest result is over a year old. | $0.57 | $2.30 | 131K | 52.3Fair value (52.3). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | 141.3Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
GLM 4.7 Flash 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. Newer version | 116.1From 186 benchmarks. Plenty of data behind it. | 83 ±10Coding IQ 82.8, 95% interval 73–93. 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. | —Not measured | —Not measured | |
GLM 4.7 Flash 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. Newer version | 116.1From 186 benchmarks. Plenty of data behind it. | 83 ±10Coding IQ 82.8, 95% interval 73–93. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.0735cache read $0.0105 | $0.42 | —Not published by the provider | 44.1Poor value (44.1). Costs more per task than its benchmark results predict. A fair amount of data behind it. | —Not measured | —Not measured | |
GLM 4.7 Flash 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. Newer version | 116.1From 186 benchmarks. Plenty of data behind it. | 83 ±10Coding IQ 82.8, 95% interval 73–93. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.0735cache read $0.0105 | $0.42 | —Not published by the provider | 44.1Poor value (44.1). Costs more per task than its benchmark results predict. A fair amount of data behind it. | —Not measured | —Not measured | |
GLM 4.5V 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 115.8From 62 benchmarks. A fair amount of data behind it. | —Not enough data yet | $0.60 | $1.80 | 66K | 36.7Poor value (36.7). Costs more per task than its benchmark results predict. Plenty of data behind it. | —Not measured | —Not measured | |
GLM 4.5V 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 115.8From 62 benchmarks. A fair amount of data behind it. | —Not enough data yet | $0.60cache read $0.11 | $1.80 | —Not published by the provider | 36.7Poor value (36.7). Costs more per task than its benchmark results predict. Plenty of data behind it. | —Not measured | —Not measured | |
GLM 4.5V 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 115.8From 62 benchmarks. A fair amount of data behind it. | —Not enough data yet | $0.60cache read $0.11 | $1.80 | 66K | 36.7Poor value (36.7). Costs more per task than its benchmark results predict. Plenty of data behind it. | —Not measured | —Not measured | |
Qwen3 32B 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 115.6From 234 benchmarks. Plenty of data behind it. | 85 ±9Coding IQ 84.8, 95% interval 76–93. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.10 | $0.45 | 41K | 35.8Poor value (35.8). Costs more per task than its benchmark results predict. A fair amount of data behind it. | —Not measured | 136.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
DeepSeek V3 🇨🇳Maker 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. | 85 ±7Coding IQ 85.3, 95% interval 79–92. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.40 | $1.30 | 64K | 43.6Poor value (43.6). Costs more per task than its benchmark results predict. Plenty of data behind it. | 47 ±9Typical (47 ±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 | |
DeepSeek V3 🇨🇳Maker 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. | 85 ±7Coding IQ 85.3, 95% interval 79–92. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.27 | $1.12 | 164K | 43.6Poor value (43.6). Costs more per task than its benchmark results predict. Plenty of data behind it. | 47 ±9Typical (47 ±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 | |
DeepSeek V3 🇨🇳Maker 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. | 85 ±7Coding IQ 85.3, 95% interval 79–92. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.89 | $0.89 | 64K | 43.6Poor value (43.6). Costs more per task than its benchmark results predict. Plenty of data behind it. | 47 ±9Typical (47 ±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 | |
Qwen3 30B A3B 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 114.4From 142 benchmarks. Plenty of data behind it. | 90 ±10Coding IQ 90.3, 95% interval 80–101. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.09 | $0.45 | 41K | 35.8Poor value (35.8). Costs more per task than its benchmark results predict. Plenty of data behind it. | —Not measured | 138.3Epoch Capabilities IndexSetting: thinkingSource: Epoch AI Benchmarking Hub API | |
GLM 4 32B 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. | 113.3From 10 benchmarks. A fair amount of data behind it. | —Not enough data yet | $0.55 | $1.66 | 32K | 58.4Good value (58.4). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate. | —Not measured | —Not measured | |
Llama 4 Maverick 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 111.8From 350 benchmarks. Plenty of data behind it. | 78 ±9Coding IQ 78.1, 95% interval 69–87. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.27 | $0.85 | 1M | 42.1Poor value (42.1). Costs more per task than its benchmark results predict. Plenty of data behind it. | 38 ±9Terse (38 ±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. | 132.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
Llama 4 Maverick 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 111.8From 350 benchmarks. Plenty of data behind it. | 78 ±9Coding IQ 78.1, 95% interval 69–87. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.27 | $0.85 | —Not published by the provider | 42.1Poor value (42.1). Costs more per task than its benchmark results predict. Plenty of data behind it. | 38 ±9Terse (38 ±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. | 132.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
Llama 4 Maverick 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 111.8From 350 benchmarks. Plenty of data behind it. | 78 ±9Coding IQ 78.1, 95% interval 69–87. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.20 | $0.85 | 1M | 42.1Poor value (42.1). Costs more per task than its benchmark results predict. Plenty of data behind it. | 38 ±9Terse (38 ±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. | 132.2Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API | |
qwen3-vl-30b 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 111.5From 81 benchmarks. Plenty of data behind it. | —Not enough data yet | $0.20 | $1.00 | 131K | 45.6Poor value (45.6). Costs more per task than its benchmark results predict. A fair amount of data behind it. | —Not measured | —Not measured | |
qwen3-vl-30b 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 111.5From 81 benchmarks. Plenty of data behind it. | —Not 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. | —Not measured | —Not measured | |
qwen3-vl-30b 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 111.5From 81 benchmarks. Plenty of data behind it. | —Not enough data yet | $0.20 | $0.70 | —Not 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. | —Not measured | —Not measured | |
qwen3-vl-30b 🇨🇳Maker headquartered in CN, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 111.5From 81 benchmarks. Plenty of data behind it. | —Not 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. | —Not measured | —Not measured | |
GPT OSS 20b 🇺🇸Maker headquartered in US, according to the company's own website. The citation is on the model page. Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo. | 111.3From 251 benchmarks. Plenty of data behind it. | 88 ±12Coding IQ 88.2, 95% interval 76–101. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ. | $0.04 | $0.15 | 131K | 48.9Fair value (48.9). Costs per task about what its benchmark results predict. Plenty of data behind it. | 40 ±9Terse (40 ±9). Generates fewer tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate. | 134.9Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API |
Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence)