Skip to content

Listings

All listings: NanoGPT

Every listing for the models on page 8 of 40 of NanoGPT, including each model's other versions and routes. Back to NanoGPT

Sold by NanoGPT

97 listings on this page sold by NanoGPT.

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
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.
119.2
From 84 benchmarks. Plenty of data behind it.
93 ±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 ±9
Terse (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.
118.9
From 111 benchmarks. Plenty of data behind it.
95 ±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.
$0.17cache read $0.085$1.53200K
57.8
Good value (57.8). Cheaper per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot 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.
118.4
From 45 benchmarks. A fair amount of data behind it.
Not enough data yet$0.30cache read $0.055$0.90128K
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
🇨🇳
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-4.6v-original · original
118.4
From 45 benchmarks. A fair amount of data behind it.
Not enough data yet$0.30cache read $0.055$0.90128K
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
GPT Image 1.5
🇺🇸
Maker headquartered in US, according to the company's own website. The citation is on the model page.
118.3
From 56 benchmarks. A fair amount of data behind it.
Not enough data yet——Not published by the providerNot enough data yetNot measuredNot measured
Sarvam 105B
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
118.3
From 14 benchmarks. A fair amount of data behind it.
Not enough data yet$0.054cache read $0.0336$0.2124131KNot enough data yetNot measuredNot 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.
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
tee
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
fp4 · GLB
Inference by Akash
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.0315$0.1964Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
Inference by Aster
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.1575$0.63Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
fast
Inference by Aster
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.1575$0.63Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
fp4 · 🇺🇸 US
Inference by Baseten
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.105$0.525Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
fp16 · 🇺🇸 US
Inference by Cerebras
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.3675$0.7875Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
fp4 · 🇺🇸 US
Inference by Coreweave
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.0315$0.1785Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
bf16 · 🇺🇸 US
Inference by Crusoe
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.0525$0.2625Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
bf16 · 🇺🇸 US
Inference by DeepInfra
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.0389cache read $0.0194$0.1785Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
bf16 · 🇮🇩 ID
Inference by Dekallm
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.0315$0.189Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
🇺🇸 US
Inference by Fireworks AI
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.1575cache read $0.0158$0.63Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
🇺🇸 US
Inference by Google
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.0945$0.378Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
🇺🇸 US
Inference by Groq
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.1575cache read $0.0788$0.63Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
mxfp4 · 🇦🇪 AE
Inference by Hyperfusion
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.0315$0.1785Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
🇪🇺 EU
Inference by Ionos
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.1785$0.7455Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
fp8
Inference by Mancer
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.0893$0.3938Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
fp4 · GLB
Inference by Nebius AI Studio
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.1575cache read $0.105$0.63Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
UNK
Inference by Ollama
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.1575cache read $0.0147$0.63Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
fp4 · 🇺🇸 US
Inference by Parasail
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.105cache read $0.0578$0.7875Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
fp4 · UNK
Inference by Primeintellect
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.3675cache read $0.3308$0.7875Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
UNK
Inference by Runware
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.0302$0.1323Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
asap · GLB
Inference by Sail Research
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.063cache read $0.0315$0.42Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
🇺🇸 US
Inference by SambaNova
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.147$0.9975Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
fp8 · GLB
Inference by Scx
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.1785$0.5775Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
fp8 · 🇸🇬 SG
Inference by SiliconFlow
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.1575cache read $0.0788$0.63Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
🇺🇸 US
Inference by Tinfoil
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.1575$0.63Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
🇺🇸 US
Inference by Together AI
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.1575$0.63Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
GLB
Inference by Uomi
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.1295$0.5375Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
int4 · 🇺🇸 US
Inference by Venice
118.2
From 424 benchmarks. Plenty of data behind it.
86 ±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.
$0.0315$0.1575Not published by the provider
58.4
Good value (58.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
37 ±11
Terse (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.
140.2
Epoch 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.
118.2
From 27 benchmarks. Plenty of data behind it.
98 ±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 ±9
Typical (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 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.
🇪🇺 EU
Inference by Ionos
118.2
From 27 benchmarks. Plenty of data behind it.
98 ±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.
$0.1785$0.9345Not published by the provider
36.8
Poor value (36.8). Costs more per task than its benchmark results predict. Plenty of data behind it.
54 ±9
Typical (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 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.
bf16 · 🇺🇸 US
Inference by Parasail
118.2
From 27 benchmarks. Plenty of data behind it.
98 ±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.
$0.126cache read $0.0735$0.84Not published by the provider
36.8
Poor value (36.8). Costs more per task than its benchmark results predict. Plenty of data behind it.
54 ±9
Typical (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
DeepSeek Chat
🇨🇳
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$0.29$1.16128K
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
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.
118.1
From 154 benchmarks. Plenty of data behind it.
92 ±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.
$0.09cache read $0.045$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
🇨🇳
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 · 🇺🇸 US
Inference by DeepInfra
118.1
From 154 benchmarks. Plenty of data behind it.
92 ±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.
$0.0945$1.155Not published by the provider
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
🇨🇳
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 · 🇺🇸 US
Inference by Parasail
118.1
From 154 benchmarks. Plenty of data behind it.
92 ±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.
$0.105cache read $0.0735$1.155Not published by the provider
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
🇨🇳
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.
93 ±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$0.85128K
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
🇨🇳
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:thinking · thinking
118.0
From 210 benchmarks. Plenty of data behind it.
93 ±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$0.85128K
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
🇨🇳
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 · 🇺🇸 US
Inference by Novita AI
118.0
From 210 benchmarks. Plenty of data behind it.
93 ±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.
$0.1365cache read $0.0263$0.8925Not 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
🇨🇳
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:thinking · thinking · bf16 · 🇺🇸 US
Inference by Novita AI
118.0
From 210 benchmarks. Plenty of data behind it.
93 ±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.
$0.1365cache read $0.0263$0.8925Not 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
🇨🇳
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 · 🇸🇬 SG
Inference by SiliconFlow
118.0
From 210 benchmarks. Plenty of data behind it.
93 ±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.
$0.147$0.903Not 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
🇨🇳
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:thinking · thinking · fp8 · 🇸🇬 SG
Inference by SiliconFlow
118.0
From 210 benchmarks. Plenty of data behind it.
93 ±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.
$0.147$0.903Not 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
🇨🇳
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 · 🇸🇬 SG
Inference by Z.ai
118.0
From 210 benchmarks. Plenty of data behind it.
93 ±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.
$0.21cache read $0.0315$1.155Not 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
🇨🇳
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:thinking · thinking · fp8 · 🇸🇬 SG
Inference by Z.ai
118.0
From 210 benchmarks. Plenty of data behind it.
93 ±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.
$0.21cache read $0.0315$1.155Not 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
Nemotron 3 Super 120B A12B
🇺🇸
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.
98 ±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.
$0.05cache read $0.025$0.25262KNot enough data yetNot measuredNot measured
Nemotron 3 Super 120B A12B
🇺🇸
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.
nvidia/nemotron-3-super-120b-a12b:thinking · thinking
117.3
From 151 benchmarks. Plenty of data behind it.
98 ±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.
$0.05cache read $0.025$0.25262KNot enough data yetNot measuredNot measured
Nemotron 3 Super 120B A12B
🇺🇸
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 · 🇺🇸 US
Inference by DeepInfra
117.3
From 151 benchmarks. Plenty of data behind it.
98 ±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.
$0.0788$0.42Not published by the providerNot enough data yetNot measuredNot measured
Nemotron 3 Super 120B A12B
🇺🇸
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.
nvidia/nemotron-3-super-120b-a12b:thinking · thinking · bf16 · 🇺🇸 US
Inference by DeepInfra
117.3
From 151 benchmarks. Plenty of data behind it.
98 ±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.
$0.0788$0.42Not published by the providerNot enough data yetNot measuredNot measured
Nemotron 3 Super 120B A12B
🇺🇸
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 · 🇮🇩 ID
Inference by Dekallm
117.3
From 151 benchmarks. Plenty of data behind it.
98 ±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.
$0.084$0.4725Not published by the providerNot enough data yetNot measuredNot measured
Nemotron 3 Super 120B A12B
🇺🇸
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.
nvidia/nemotron-3-super-120b-a12b:thinking · thinking · fp8 · 🇮🇩 ID
Inference by Dekallm
117.3
From 151 benchmarks. Plenty of data behind it.
98 ±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.
$0.084$0.4725Not published by the providerNot enough data yetNot measuredNot measured
Nemotron 3 Super 120B A12B
🇺🇸
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.
UNK
Inference by Ollama
117.3
From 151 benchmarks. Plenty of data behind it.
98 ±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.
$0.0158$0.63Not published by the providerNot enough data yetNot measuredNot measured
Nemotron 3 Super 120B A12B
🇺🇸
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.
nvidia/nemotron-3-super-120b-a12b:thinking · thinking · UNK
Inference by Ollama
117.3
From 151 benchmarks. Plenty of data behind it.
98 ±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.
$0.0158$0.63Not published by the providerNot enough data yetNot measuredNot measured
Qwen3.5 Omni Flash
🇨🇳
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
117.3
From 54 benchmarks. A fair amount of data behind it.
Not enough data yet——49KNot enough data yetNot measuredNot measured
Codestral 25.01
🇫🇷
Maker headquartered in FR, according to the company's own website. The citation is on the model page.
mistralai/codestral-2508 · 2508
116.6
From 10 benchmarks. A fair amount of data behind it.
Not enough data yet$0.30cache read $0.15$0.90256K
37.8
Poor value (37.8). Costs more per task than its benchmark results predict. A fair amount of data behind it.
40 ±9
Terse (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.
Not measured
Mercury 2
Newer version
Newer versionMercury 3We compared release dates and version names in our catalogue.
116.5
From 53 benchmarks. Plenty of data behind it.
93 ±13
Coding IQ 93.2, 95% interval 80–106. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.25cache read $0.025$0.75128K
56.7
Good value (56.7). Cheaper per task than its benchmark results predict. A fair amount of data behind it.
43 ±9
Terse (43 ±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 Medium 3
🇫🇷
Maker headquartered in FR, according to the company's own website. The citation is on the model page.
Newer version
Newer versionMistral Medium 3.5We compared release dates and version names in our catalogue.
116.4
From 50 benchmarks. A fair amount of data behind it.
Not enough data yet$0.44cache read $0.044$2.20131K
51.0
Fair value (51.0). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measured
132.8
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Qwen Plus
🇨🇳
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
116.4
From 42 benchmarks. Plenty of data behind it.
Not enough data yet$0.3995cache read $0.1998$1.2002996K
43.6
Poor value (43.6). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Nemotron 3.5 Lightning
🇺🇸
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.
116.2
From 12 benchmarks. A fair amount of data behind it.
Not enough data yet$0.05cache read $0.01$0.20262K
49.3
Fair value (49.3). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Nemotron 3.5 Lightning
🇺🇸
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-3.5-lightning:thinking · thinking
116.2
From 12 benchmarks. A fair amount of data behind it.
Not enough data yet$0.05cache read $0.01$0.20262K
49.3
Fair value (49.3). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Nemotron 3.5 Lightning
🇺🇸
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 · 🇺🇸 US
Inference by Coreweave
116.2
From 12 benchmarks. A fair amount of data behind it.
Not enough data yet$0.0735cache read $0.042$0.21Not published by the provider
49.3
Fair value (49.3). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Nemotron 3.5 Lightning
🇺🇸
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-3.5-lightning:thinking · thinking · bf16 · 🇺🇸 US
Inference by Coreweave
116.2
From 12 benchmarks. A fair amount of data behind it.
Not enough data yet$0.0735cache read $0.042$0.21Not published by the provider
49.3
Fair value (49.3). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Nemotron 3.5 Lightning
🇺🇸
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 · GLB
Inference by Darkbloom
116.2
From 12 benchmarks. A fair amount of data behind it.
Not enough data yet$0.021$0.1008Not published by the provider
49.3
Fair value (49.3). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Nemotron 3.5 Lightning
🇺🇸
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-3.5-lightning:thinking · thinking · int4 · GLB
Inference by Darkbloom
116.2
From 12 benchmarks. A fair amount of data behind it.
Not enough data yet$0.021$0.1008Not published by the provider
49.3
Fair value (49.3). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Nemotron 3.5 Lightning
🇺🇸
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 · 🇺🇸 US
Inference by DeepInfra
116.2
From 12 benchmarks. A fair amount of data behind it.
Not enough data yet$0.063cache read $0.0315$0.168Not published by the provider
49.3
Fair value (49.3). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Nemotron 3.5 Lightning
🇺🇸
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-3.5-lightning:thinking · thinking · bf16 · 🇺🇸 US
Inference by DeepInfra
116.2
From 12 benchmarks. A fair amount of data behind it.
Not enough data yet$0.063cache read $0.0315$0.168Not published by the provider
49.3
Fair value (49.3). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Nemotron 3.5 Lightning
🇺🇸
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.
🇺🇸 US
Inference by Fireworks AI
116.2
From 12 benchmarks. A fair amount of data behind it.
Not enough data yet$0.0525cache read $0.0105$0.21Not published by the provider
49.3
Fair value (49.3). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Nemotron 3.5 Lightning
🇺🇸
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-3.5-lightning:thinking · thinking · 🇺🇸 US
Inference by Fireworks AI
116.2
From 12 benchmarks. A fair amount of data behind it.
Not enough data yet$0.0525cache read $0.0105$0.21Not published by the provider
49.3
Fair value (49.3). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Nemotron 3.5 Lightning
🇺🇸
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
Inference by Modelrun
116.2
From 12 benchmarks. A fair amount of data behind it.
Not enough data yet$0.0735cache read $0.042$0.21Not published by the provider
49.3
Fair value (49.3). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Nemotron 3.5 Lightning
🇺🇸
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-3.5-lightning:thinking · thinking · fp4 · 🇺🇸 US
Inference by Modelrun
116.2
From 12 benchmarks. A fair amount of data behind it.
Not enough data yet$0.0735cache read $0.042$0.21Not published by the provider
49.3
Fair value (49.3). 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 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.
116.2
From 18 benchmarks. A fair amount of data behind it.
Not enough data yet$0.10cache read $0.05$0.40128K
54.3
Good value (54.3). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot 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
Newer versionGLM 5.3 FlashWe compared release dates and version names in our catalogue.
116.1
From 186 benchmarks. Plenty of data behind it.
83 ±10
Coding 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.
$0.0605cache read $0.0303$0.40200K
44.1
Poor value (44.1). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot 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
Newer versionGLM 5.3 FlashWe compared release dates and version names in our catalogue.
z-ai/glm-4.7-flash:thinking · thinking
116.1
From 186 benchmarks. Plenty of data behind it.
83 ±10
Coding 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.
$0.0605cache read $0.0303$0.40200K
44.1
Poor value (44.1). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot 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
Newer versionGLM 5.3 FlashWe compared release dates and version names in our catalogue.
🇺🇸 US
Inference by Cloudflare Workers AI
116.1
From 186 benchmarks. Plenty of data behind it.
83 ±10
Coding 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.
$0.0635$0.42Not published by the provider
44.1
Poor value (44.1). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot 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
Newer versionGLM 5.3 FlashWe compared release dates and version names in our catalogue.
z-ai/glm-4.7-flash:thinking · thinking · 🇺🇸 US
Inference by Cloudflare Workers AI
116.1
From 186 benchmarks. Plenty of data behind it.
83 ±10
Coding 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.
$0.0635$0.42Not published by the provider
44.1
Poor value (44.1). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot 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
Newer versionGLM 5.3 FlashWe compared release dates and version names in our catalogue.
fp8 · 🇪🇺 EU
Inference by Greenference
116.1
From 186 benchmarks. Plenty of data behind it.
83 ±10
Coding 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 $0.015$0.4279Not published by the provider
44.1
Poor value (44.1). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot 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
Newer versionGLM 5.3 FlashWe compared release dates and version names in our catalogue.
z-ai/glm-4.7-flash:thinking · thinking · fp8 · 🇪🇺 EU
Inference by Greenference
116.1
From 186 benchmarks. Plenty of data behind it.
83 ±10
Coding 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 $0.015$0.4279Not published by the provider
44.1
Poor value (44.1). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot 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
Newer versionGLM 5.3 FlashWe compared release dates and version names in our catalogue.
bf16 · 🇺🇸 US
Inference by Novita AI
116.1
From 186 benchmarks. Plenty of data behind it.
83 ±10
Coding 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$0.42Not published by the provider
44.1
Poor value (44.1). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot 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
Newer versionGLM 5.3 FlashWe compared release dates and version names in our catalogue.
z-ai/glm-4.7-flash:thinking · thinking · bf16 · 🇺🇸 US
Inference by Novita AI
116.1
From 186 benchmarks. Plenty of data behind it.
83 ±10
Coding 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$0.42Not published by the provider
44.1
Poor value (44.1). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measuredNot 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.
115.8
From 62 benchmarks. A fair amount of data behind it.
Not enough data yet$0.60cache read $0.11$1.8066K
36.7
Poor value (36.7). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measuredNot 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.
z-ai/glm-4.5v:thinking · thinking
115.8
From 62 benchmarks. A fair amount of data behind it.
Not enough data yet$0.60cache read $0.11$1.8066K
36.7
Poor value (36.7). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measuredNot measured
Gemini 3.1 Flash Image
🇺🇸
Maker headquartered in US, according to the company's own website. The citation is on the model page.
nano-banana-2
115.7
From 40 benchmarks. A fair amount of data behind it.
Not enough data yet——Not published by the providerNot enough data yetNot measuredNot measured
Ministral 14B
🇫🇷
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/ministral-14b-2512 · 2512
115.6
From 33 benchmarks. A fair amount of data behind it.
Not enough data yet$0.20cache read $0.02$0.20262KNot enough data yetNot measuredNot measured
Ministral 14B
🇫🇷
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/ministral-14b-2512 · 2512 · int4 · GLB
Inference by Kitani
115.6
From 33 benchmarks. A fair amount of data behind it.
Not enough data yet$0.189cache read $0.0945$0.189Not published by the providerNot enough data yetNot measuredNot 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.
115.6
From 234 benchmarks. Plenty of data behind it.
85 ±9
Coding 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.
$0.08$0.2841K
35.8
Poor value (35.8). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measured
136.9
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
fp8 · 🇺🇸 US
Inference by DeepInfra
115.6
From 234 benchmarks. Plenty of data behind it.
85 ±9
Coding 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.
$0.084$0.294Not published by the provider
35.8
Poor value (35.8). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measured
136.9
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
fp8 · 🇪🇺 EU
Inference by Greenference
115.6
From 234 benchmarks. Plenty of data behind it.
85 ±9
Coding 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.
$0.0998cache read $0.0499$0.2993Not published by the provider
35.8
Poor value (35.8). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measured
136.9
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
bf16 · GLB
Inference by Scx
115.6
From 234 benchmarks. Plenty of data behind it.
85 ±9
Coding 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.
$0.378$0.9135Not published by the provider
35.8
Poor value (35.8). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measured
136.9
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
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.
fp8 · 🇸🇬 SG
Inference by SiliconFlow
115.6
From 234 benchmarks. Plenty of data behind it.
85 ±9
Coding 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.
$0.147$0.5985Not published by the provider
35.8
Poor value (35.8). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measured
136.9
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT-4.1 mini
🇺🇸
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Newer version
Newer versionGPT-5.4 miniWe compared the version names in our catalogue.
115.4
From 375 benchmarks. Plenty of data behind it.
91 ±8
Coding IQ 91.3, 95% interval 84–99. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.40cache read $0.10$1.601M
46.2
Poor value (46.2). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measured
134.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GPT-5 nano
🇺🇸
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Newer version
Newer versionGPT 5.4 NanoWe compared release dates and version names in our catalogue.
115.3
From 332 benchmarks. Plenty of data behind it.
93 ±13
Coding IQ 92.7, 95% interval 80–106. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.05cache read $0.005$0.40400K
46.3
Poor value (46.3). Costs more per task than its benchmark results predict. Plenty of data behind it.
51 ±9
Typical (51 ±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
139.1
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API

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

How this data works