Skip to content

Listings

All listings: OpenRouter

Every listing for the models on page 9 of 26 of OpenRouter, including each model's other versions and routes. Back to OpenRouter

Sold by OpenRouter

82 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
Modelllminfo IQCoding IQInput $/1MOutput $/1MMax Input TokensValueYapEpoch Capabilities Index
Claude 3.5 Sonnet
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Newer version
Newer versionClaude Sonnet 5.5We compared release dates and version names in our catalogue.
119.3
From 546 benchmarks. Plenty of data behind it.
89Β Β±9
Coding IQ 89.4, 95% interval 81–98. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$3.00cache read $0.30$15.00cache write $3.75Not published by the provider
27.0
Poor value (27.0). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measured
130
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Claude 3.5 Sonnet
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Newer version
Newer versionClaude Sonnet 5.5We compared release dates and version names in our catalogue.
anthropic/claude-3.5-sonnet-20240620 Β· 20240620
119.3
From 546 benchmarks. Plenty of data behind it.
89Β Β±9
Coding IQ 89.4, 95% interval 81–98. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$3.00$15.00Not published by the provider
27.0
Poor value (27.0). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measured
130
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Codex Mini
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
β‰ˆ 119
From 9 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet$1.50cache read $0.375$6.00Not published by the provider
39.7
Poor value (39.7). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot measured
Codex Mini
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
openai/codex-mini-latest Β· latest
β‰ˆ 119
From 9 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet$1.50$6.00Not published by the provider
39.7
Poor value (39.7). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot measured
Qwen 3 235B A22B
πŸ‡¨πŸ‡³
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.
119.3
From 185 benchmarks. Plenty of data behind it.
86Β Β±12
Coding 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.
$0.13$0.60Not published by the provider
53.2
Good value (53.2). Cheaper per task than its benchmark results predict. Plenty of data behind it.
Not measured
143.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Qwen 3 235B A22B
πŸ‡¨πŸ‡³
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.
qwen/qwen3-235b-a22b:free Β· free
119.3
From 185 benchmarks. Plenty of data behind it.
86Β Β±12
Coding 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.
$0.00$0.00Not published by the provider
53.2
Good value (53.2). Cheaper per task than its benchmark results predict. Plenty of data behind it.
Not measured
143.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
MiMo V2 Flash
πŸ‡¨πŸ‡³
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
Newer versionMiMo V2.6 FlashWe compared release dates and version names in our catalogue.
119.2
From 153 benchmarks. Plenty of data behind it.
95Β Β±11
Coding 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.
$0.10cache read $0.01$0.30Not published by the provider
53.7
Good value (53.7). Cheaper per task than its benchmark results predict. A fair amount of data behind it.
53 Β±9
Typical (53 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.
Not measured
Qwen3 Coder
πŸ‡¨πŸ‡³
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.
qwen/qwen3-coder:free Β· free
119.2
From 84 benchmarks. Plenty of data behind it.
93Β Β±6
Coding 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.
$0.00$0.00Not published by the provider
34.9
Poor value (34.9). Costs more per task than its benchmark results predict. Plenty of data behind it.
39 Β±9
Terse (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.
Not measured
Qwen3 Coder
πŸ‡¨πŸ‡³
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.
119.2
From 84 benchmarks. Plenty of data behind it.
93Β Β±6
Coding 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.
$0.30cache read $0.10$1.00262K
34.9
Poor value (34.9). Costs more per task than its benchmark results predict. Plenty of data behind it.
39 Β±9
Terse (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.
Not measured
Qwen3 Coder
πŸ‡¨πŸ‡³
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.
turbo Β· fp4
Inference by DeepInfra
119.2
From 84 benchmarks. Plenty of data behind it.
93Β Β±6
Coding 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.
$0.30cache read $0.10$1.00262K
34.9
Poor value (34.9). Costs more per task than its benchmark results predict. Plenty of data behind it.
39 Β±9
Terse (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.
Not measured
MiniMax M2
πŸ‡¨πŸ‡³
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.
118.9
From 111 benchmarks. Plenty of data behind it.
95Β Β±7
Coding 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.
$0.255cache read $0.03$1.00Not published by the provider
57.8
Good value (57.8). Cheaper per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot measured
Hermes 4 405B
πŸ‡ΊπŸ‡Έ
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.
118.8
From 7 benchmarks. A fair amount of data behind it.
Not enough data yet$1.00cache read $0.20$3.00131K
38.3
Poor value (38.3). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Hermes 4 405B
πŸ‡ΊπŸ‡Έ
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.
fp8
Inference by Nebius AI Studio
118.8
From 7 benchmarks. A fair amount of data behind it.
Not enough data yet$1.00cache read $0.20$3.00131K
38.3
Poor value (38.3). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Nemotron 3 Nano Omni 30b A3b Reasoning
πŸ‡ΊπŸ‡Έ
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.
nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free Β· free
β‰ˆ 119
From 3 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet$0.00$0.00256KNot enough data yetNot measuredNot measured
Nemotron 3 Nano Omni 30b A3b Reasoning
πŸ‡ΊπŸ‡Έ
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.
nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free Β· free
Inference by NVIDIA NIM
β‰ˆ 119
From 3 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet$0.00$0.00256KNot enough data yetNot measuredNot measured
118.6
From 76 benchmarks. Plenty of data behind it.
Not enough data yet$0.10$0.40262K
45.0
Poor value (45.0). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Inference by Seed
118.6
From 76 benchmarks. Plenty of data behind it.
Not enough data yet$0.10$0.40262K
45.0
Poor value (45.0). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
GLM 4.6v
πŸ‡¨πŸ‡³
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.
118.4
From 45 benchmarks. A fair amount of data behind it.
Not enough data yet$0.30cache read $0.055$0.90131K
44.9
Poor value (44.9). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot measured
GLM 4.6v
πŸ‡¨πŸ‡³
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.
bf16
Inference by Novita AI
118.4
From 45 benchmarks. A fair amount of data behind it.
Not enough data yet$0.30cache read $0.055$0.90131K
44.9
Poor value (44.9). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot measured
GLM 4.6v
πŸ‡¨πŸ‡³
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.
fp8
Inference by Z.ai
118.4
From 45 benchmarks. A fair amount of data behind it.
Not enough data yet$0.30cache read $0.05$0.90131K
44.9
Poor value (44.9). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot measured
Grok Code Fast 1
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
118.4
From 31 benchmarks. Plenty of data behind it.
89Β Β±8
Coding IQ 88.7, 95% interval 81–96. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.20cache read $0.02$1.50256K
52.1
Fair value (52.1). Costs per task about what its benchmark results predict. A fair amount of data behind it.
43 Β±9
Terse (43 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.
Not measured
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.09$0.36131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
openai/gpt-oss-120b:batch Β· batch
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.0296$0.136131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
fp4
Inference by Akash
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.03$0.187131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
Inference by Amazon Web Services
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.15$0.60131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
eu-west-1
Inference by Amazon Web Services
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.15$0.60131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
fp4
Inference by Baseten
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.10$0.50128K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
fp4/us Β· fp4
Inference by Baseten
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.10$0.50128K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
fp16
Inference by Cerebras
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.35$0.75131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
fp4
Inference by Coreweave
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.03$0.17131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
bf16
Inference by Crusoe
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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$0.25131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
openai/gpt-oss-120b:batch Β· batch+bf16 Β· bf16
Inference by DeepInfra
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.0296$0.136131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
bf16
Inference by DeepInfra
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.037cache read $0.0185$0.17131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
Inference by Digitalocean
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.06cache read $0.012$0.42128K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
Inference by Fireworks AI
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.15cache read $0.015$0.60131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
global
Inference by Google
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.09$0.36131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
Inference by Groq
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.15cache read $0.075$0.60131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
fp8
Inference by Mancer 2
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.085$0.375131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
fp4
Inference by Near AI
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.03cache read $0.015$0.15131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
fast Β· fp4
Inference by Parasail
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.15$0.75131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
fp4
Inference by Parasail
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.10cache read $0.055$0.75131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
fp4
Inference by Primeintellect
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.35cache read $0.315$0.75131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
Inference by SambaNova
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.14$0.95131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
fp8
Inference by SiliconFlow
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.15cache read $0.075$0.60131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
Inference by Together AI
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.15$0.60131K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT OSS 120B
πŸ‡ΊπŸ‡Έ
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.
int4
Inference by Venice
118.2
From 424 benchmarks. Plenty of data behind it.
86Β Β±5
Coding 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.
$0.03$0.15128K
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 Β±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Grok 3
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
118.2
From 188 benchmarks. Plenty of data behind it.
β‰ˆ 91Β Β±10stale
Coding IQ 90.7, 95% interval 80–101. 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.
$3.00cache read $0.75$15.00Not published by the provider
34.0
Poor value (34.0). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measured
137.6
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Grok 3
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
x-ai/grok-3-beta Β· beta
118.2
From 188 benchmarks. Plenty of data behind it.
β‰ˆ 91Β Β±10stale
Coding IQ 90.7, 95% interval 80–101. 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.
$3.00$15.00Not published by the provider
34.0
Poor value (34.0). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measured
137.6
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
LongCat Flash Chat
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
118.2
From 58 benchmarks. Plenty of data behind it.
94Β Β±12
Coding IQ 94.3, 95% interval 82–106. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.15$0.75Not published by the provider
47.8
Fair value (47.8). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
LongCat Flash Chat
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
meituan/longcat-flash-chat:free Β· free
118.2
From 58 benchmarks. Plenty of data behind it.
94Β Β±12
Coding IQ 94.3, 95% interval 82–106. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.00$0.00Not published by the provider
47.8
Fair value (47.8). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Qwen3 Coder Next
πŸ‡¨πŸ‡³
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.
118.2
From 27 benchmarks. Plenty of data behind it.
98Β Β±7
Coding 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.
$0.12cache read $0.07$0.80262K
36.8
Poor value (36.8). Costs more per task than its benchmark results predict. Plenty of data behind it.
54 Β±9
Typical (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.
Not measured
Qwen3 Coder Next
πŸ‡¨πŸ‡³
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.
bf16
Inference by Parasail
118.2
From 27 benchmarks. Plenty of data behind it.
98Β Β±7
Coding 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.
$0.12cache read $0.07$0.80262K
36.8
Poor value (36.8). Costs more per task than its benchmark results predict. Plenty of data behind it.
54 Β±9
Typical (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.
Not measured
DeepSeek Chat
πŸ‡¨πŸ‡³
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.
118.1
From 32 benchmarks. Plenty of data behind it.
Not enough data yet$0.2574$1.0287128K
45.0
Poor value (45.0). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot measured
DeepSeek Chat
πŸ‡¨πŸ‡³
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.
fp4
Inference by DeepInfra
118.1
From 32 benchmarks. Plenty of data behind it.
Not enough data yet$0.32$0.89164K
45.0
Poor value (45.0). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot measured
DeepSeek Chat
πŸ‡¨πŸ‡³
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.
fp4/flex Β· fp4
Inference by DeepInfra
118.1
From 32 benchmarks. Plenty of data behind it.
Not enough data yet$0.256$0.712164K
45.0
Poor value (45.0). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot measured
DeepSeek Chat
πŸ‡¨πŸ‡³
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.
Inference by Streamlake
118.1
From 32 benchmarks. Plenty of data behind it.
Not enough data yet$0.2574$1.0287128K
45.0
Poor value (45.0). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot measured
Mistral Large 3
πŸ‡«πŸ‡·
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.
mistralai/mistral-large-2512
118.1
From 60 benchmarks. Plenty of data behind it.
86Β Β±8
Coding IQ 86.4, 95% interval 79–94. 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$1.50262K
38.5
Poor value (38.5). Costs more per task than its benchmark results predict. Plenty of data behind it.
40 Β±9
Terse (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.
Not measured
Mistral Large 3
πŸ‡«πŸ‡·
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.
mistralai/mistral-large-2512 Β· eu
Inference by Mistral AI
118.1
From 60 benchmarks. Plenty of data behind it.
86Β Β±8
Coding IQ 86.4, 95% interval 79–94. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.55cache read $0.055$1.65262K
38.5
Poor value (38.5). Costs more per task than its benchmark results predict. Plenty of data behind it.
40 Β±9
Terse (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.
Not measured
Mistral Large 3
πŸ‡«πŸ‡·
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.
mistralai/mistral-large-2512 Β· zdr
Inference by Mistral AI
118.1
From 60 benchmarks. Plenty of data behind it.
86Β Β±8
Coding IQ 86.4, 95% interval 79–94. 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$1.50262K
38.5
Poor value (38.5). Costs more per task than its benchmark results predict. Plenty of data behind it.
40 Β±9
Terse (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.
Not measured
Qwen3 Next 80B A3B Instruct
πŸ‡¨πŸ‡³
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.
118.1
From 154 benchmarks. Plenty of data behind it.
92Β Β±11
Coding 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.
$0.10cache read $0.07$1.10262K
47.6
Fair value (47.6). Costs per task about what its benchmark results predict. A fair amount of data behind it.
Not measuredNot measured
Qwen3 Next 80B A3B Instruct
πŸ‡¨πŸ‡³
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.
fp8
Inference by DeepInfra
118.1
From 154 benchmarks. Plenty of data behind it.
92Β Β±11
Coding 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.
$0.09$1.10262K
47.6
Fair value (47.6). Costs per task about what its benchmark results predict. A fair amount of data behind it.
Not measuredNot measured
Qwen3 Next 80B A3B Instruct
πŸ‡¨πŸ‡³
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.
fp8/flex Β· fp8
Inference by DeepInfra
118.1
From 154 benchmarks. Plenty of data behind it.
92Β Β±11
Coding 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.
$0.072$0.88262K
47.6
Fair value (47.6). Costs per task about what its benchmark results predict. A fair amount of data behind it.
Not measuredNot measured
Qwen3 Next 80B A3B Instruct
πŸ‡¨πŸ‡³
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.
fp8/priority Β· fp8
Inference by DeepInfra
118.1
From 154 benchmarks. Plenty of data behind it.
92Β Β±11
Coding 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.
$0.135$1.65262K
47.6
Fair value (47.6). Costs per task about what its benchmark results predict. A fair amount of data behind it.
Not measuredNot measured
Qwen3 Next 80B A3B Instruct
πŸ‡¨πŸ‡³
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.
fp8
Inference by Parasail
118.1
From 154 benchmarks. Plenty of data behind it.
92Β Β±11
Coding 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.
$0.10cache read $0.07$1.10262K
47.6
Fair value (47.6). Costs per task about what its benchmark results predict. A fair amount of data behind it.
Not measuredNot measured
GLM 4.5 Air
πŸ‡¨πŸ‡³
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.
z-ai/glm-4.5-air:free Β· free
118.0
From 210 benchmarks. Plenty of data behind it.
93Β Β±7
Coding 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.
$0.00$0.00Not published by the provider
42.6
Poor value (42.6). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measuredNot measured
GLM 4.5 Air
πŸ‡¨πŸ‡³
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.
118.0
From 210 benchmarks. Plenty of data behind it.
93Β Β±7
Coding 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$0.85131K
42.6
Poor value (42.6). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measuredNot measured
GLM 4.5 Air
πŸ‡¨πŸ‡³
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.
bf16
Inference by Novita AI
118.0
From 210 benchmarks. Plenty of data behind it.
93Β Β±7
Coding 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$0.85131K
42.6
Poor value (42.6). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measuredNot measured
GLM 4.5 Air
πŸ‡¨πŸ‡³
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.
fp8
Inference by SiliconFlow
118.0
From 210 benchmarks. Plenty of data behind it.
93Β Β±7
Coding 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.
$0.14$0.86131K
42.6
Poor value (42.6). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measuredNot measured
GLM 4.5 Air
πŸ‡¨πŸ‡³
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.
fp8
Inference by Z.ai
118.0
From 210 benchmarks. Plenty of data behind it.
93Β Β±7
Coding 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.
$0.20cache read $0.03$1.10131K
42.6
Poor value (42.6). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measuredNot measured
Kimi K2
πŸ‡¨πŸ‡³
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
Newer versionKimi K3.1We compared the version names in our catalogue.
moonshotai/kimi-k2:free Β· free
117.4
From 30 benchmarks. Plenty of data behind it.
β‰ˆ 95Β Β±16stale
Coding 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.
$0.00$0.00Not published by the provider
52.3
Fair 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
141.3
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Kimi K2
πŸ‡¨πŸ‡³
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
Newer versionKimi K3.1We compared the version names in our catalogue.
117.4
From 30 benchmarks. Plenty of data behind it.
β‰ˆ 95Β Β±16stale
Coding 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$2.30131K
52.3
Fair 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
141.3
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Kimi K2
πŸ‡¨πŸ‡³
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
Newer versionKimi K3.1We compared the version names in our catalogue.
fp8
Inference by Novita AI
117.4
From 30 benchmarks. Plenty of data behind it.
β‰ˆ 95Β Β±16stale
Coding 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$2.30131K
52.3
Fair 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
141.3
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Nemotron 3 Super 120B A12B
πŸ‡ΊπŸ‡Έ
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.
Newer version
Newer versionNemotron 3.5 Super 120B A12BWe compared the version names in our catalogue.
117.3
From 151 benchmarks. Plenty of data behind it.
98Β Β±20
Coding IQ 98.4, 95% interval 79–118. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.075$0.40262KNot enough data yetNot measuredNot measured
Nemotron 3 Super 120B A12B
πŸ‡ΊπŸ‡Έ
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.
Newer version
Newer versionNemotron 3.5 Super 120B A12BWe compared the version names in our catalogue.
bf16
Inference by DeepInfra
117.3
From 151 benchmarks. Plenty of data behind it.
98Β Β±20
Coding IQ 98.4, 95% interval 79–118. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.075$0.40262KNot enough data yetNot measuredNot measured
Nemotron 3 Super 120B A12B
πŸ‡ΊπŸ‡Έ
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.
Newer version
Newer versionNemotron 3.5 Super 120B A12BWe compared the version names in our catalogue.
bf16/flex Β· bf16
Inference by DeepInfra
117.3
From 151 benchmarks. Plenty of data behind it.
98Β Β±20
Coding IQ 98.4, 95% interval 79–118. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.06$0.32262KNot enough data yetNot measuredNot measured
Nemotron 3 Super 120B A12B
πŸ‡ΊπŸ‡Έ
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.
Newer version
Newer versionNemotron 3.5 Super 120B A12BWe compared the version names in our catalogue.
bf16/priority Β· bf16
Inference by DeepInfra
117.3
From 151 benchmarks. Plenty of data behind it.
98Β Β±20
Coding IQ 98.4, 95% interval 79–118. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.1125$0.60262KNot enough data yetNot measuredNot measured
Nemotron 3 Super 120B A12B
πŸ‡ΊπŸ‡Έ
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.
Newer version
Newer versionNemotron 3.5 Super 120B A12BWe compared the version names in our catalogue.
fp8
Inference by Dekallm
117.3
From 151 benchmarks. Plenty of data behind it.
98Β Β±20
Coding IQ 98.4, 95% interval 79–118. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.08$0.45262KNot enough data yetNot measuredNot measured
Solar Pro 3
πŸ‡°πŸ‡·
Maker headquartered in KR, according to the company's own website. The citation is on the model page.
Newer version
Newer versionSolar Pro 4We compared release dates and version names in our catalogue.
β‰ˆ 117
From 3 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet$0.15cache read $0.015$0.60Not published by the providerNot enough data yetNot measuredNot measured
Devstral Medium
πŸ‡«πŸ‡·
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‰ˆ 117
From 10 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet$0.40$2.00Not published by the provider
42.0
Poor value (42.0). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
β‰ˆ 117
From 41 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet$0.20$1.10Not published by the provider
40.8
Poor value (40.8). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
MiniMax M1
πŸ‡¨πŸ‡³
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
Newer versionMiniMax M2We compared release dates and version names in our catalogue.
β‰ˆ 117
From 44 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet$0.30$1.65Not published by the provider
44.8
Poor value (44.8). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
MiniMax M1
πŸ‡¨πŸ‡³
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
Newer versionMiniMax M2We compared release dates and version names in our catalogue.
minimax-m1:extended Β· extended
β‰ˆ 117
From 44 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet$0.55$2.20Not published by the provider
44.8
Poor value (44.8). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured

Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence)

How this data works