Skip to content

Listings

All listings: DeepInfra

Every listing for the models on page 1 of 11 of DeepInfra, including each model's other versions and routes. Back to DeepInfra

Sold by DeepInfra

24 listings on this page sold by DeepInfra.

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 Opus 5.5
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
146.6
From 329 benchmarks. Plenty of data behind it.
147Β Β±5
Coding IQ 147.1, 95% interval 142–152. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
β€”β€”Not published by the provider
66.8
Good value (66.8). Cheaper per task than its benchmark results predict. Plenty of data behind it.
65 Β±12
Chatty (65 Β±12). Generates more tokens than models with similar coding results. Measured at max effort. A fair amount of data behind it.
168.7
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Qwen4 Max
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
New
Released 11 Nov 2026
144.9
From 40 benchmarks. Plenty of data behind it.
β‰ˆ 124Β Β±12
Coding IQ 124.3, 95% interval 112–137. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$1.413cache read $0.071$4.241M
61.7
Good value (61.7). Cheaper per task than its benchmark results predict. Plenty of data behind it.
Not measuredNot measured
Claude Opus 5
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 Opus 5.5We compared release dates and version names in our catalogue.
143.9
From 389 benchmarks. Plenty of data behind it.
134Β Β±6
Coding IQ 133.8, 95% interval 128–140. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$5.00$25.001M
64.6
Good value (64.6). Cheaper per task than its benchmark results predict. Plenty of data behind it.
61 Β±9
Chatty (61 Β±9). Generates more tokens than models with similar coding results. Measured at max effort. A fair amount of data behind it.
164.7
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Claude Sonnet 5.5
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
143.9
From 270 benchmarks. Plenty of data behind it.
140Β Β±5
Coding IQ 140.0, 95% interval 135–145. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
β€”β€”Not published by the provider
66.7
Good value (66.7). Cheaper per task than its benchmark results predict. Plenty of data behind it.
38 Β±9
Terse (38 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.
166.6
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Claude Fable 5
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 Fable 5.2We compared release dates and version names in our catalogue.
143.0
From 383 benchmarks. Plenty of data behind it.
132Β Β±5
Coding IQ 132.1, 95% interval 128–137. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$10.00$50.001M
57.1
Good value (57.1). Cheaper per task than its benchmark results predict. Plenty of data behind it.
53 Β±10
Typical (53 Β±10). Generates about as many tokens as models with similar coding results. Measured at default effort. A fair amount of data behind it.
164.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Kimi K3.1 Pro
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.
142.7
From 161 benchmarks. Plenty of data behind it.
131Β Β±10
Coding IQ 131.3, 95% interval 121–141. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$2.00cache read $0.20$10.001M
49.2
Fair value (49.2). Costs per task about what its benchmark results predict. Plenty of data behind it.
51 Β±9
Typical (51 Β±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
161.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Claude Opus 4.8
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 Opus 5.5We compared release dates and version names in our catalogue.
139.7
From 478 benchmarks. Plenty of data behind it.
121Β Β±4
Coding IQ 120.8, 95% interval 116–125. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$5.00$25.001M
52.9
Good value (52.9). Cheaper per task than its benchmark results predict. Plenty of data behind it.
40 Β±10
Terse (40 Β±10). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it.
159.9
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4.1 Flash 1017
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
139.7
From 97 benchmarks. Plenty of data behind it.
127Β Β±8
Coding IQ 126.8, 95% interval 119–135. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.14cache read $0.0042$0.421M
71.4
Good value (71.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
55 Β±9
Typical (55 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.
159.4
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Kimi K3
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.
mxfp4
139.2
From 417 benchmarks. Plenty of data behind it.
124Β Β±7
Coding IQ 123.8, 95% interval 117–131. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$2.85cache read $0.285$14.25cache write $3.56251M
60.2
Good value (60.2). Cheaper per task than its benchmark results predict. Plenty of data behind it.
55 Β±9
Typical (55 Β±9). Generates about as many tokens as models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate.
158.6
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Gemini 3.7 Flash
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Newer version
Newer versionGemini 3.8 FlashWe compared release dates and version names in our catalogue.
138.8
From 267 benchmarks. Plenty of data behind it.
121Β Β±9
Coding IQ 121.3, 95% interval 113–130. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.75$3.751M
60.3
Good value (60.3). Cheaper per task than its benchmark results predict. Plenty of data behind it.
59 Β±12
Chatty (59 Β±12). Generates more tokens than models with similar coding results. Measured at high effort. A fair amount of data behind it.
158.4
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Qwen3.8 Max
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
Newer version
Newer versionQwen4 MaxWe compared release dates and version names in our catalogue.
138.7
From 312 benchmarks. Plenty of data behind it.
117Β Β±5
Coding IQ 117.2, 95% interval 112–122. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$1.65cache read $0.206$4.951256K
49.2
Fair value (49.2). Costs per task about what its benchmark results predict. Plenty of data behind it.
62 Β±9
Chatty (62 Β±9). Generates more tokens than models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate.
157.6
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Claude Opus 4.7
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 Opus 5.5We compared release dates and version names in our catalogue.
138.2
From 354 benchmarks. Plenty of data behind it.
118Β Β±15
Coding IQ 118.2, 95% interval 103–134. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$5.00$25.001M
49.6
Fair value (49.6). Costs per task about what its benchmark results predict. Plenty of data behind it.
38 Β±12
Terse (38 Β±12). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it.
157.5
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GLM 5.3
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
138.2
From 318 benchmarks. Plenty of data behind it.
121Β Β±9
Coding IQ 121.0, 95% interval 112–130. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.90cache read $0.20$4.001M
57.7
Good value (57.7). Cheaper per task than its benchmark results predict. Plenty of data behind it.
65 Β±10
Chatty (65 Β±10). Generates more tokens than models with similar coding results. Measured at max effort. A fair amount of data behind it.
157.1
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4.1 Flash
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
fp8
137.7
From 179 benchmarks. Plenty of data behind it.
125Β Β±16
Coding IQ 124.7, 95% interval 109–140. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.14cache read $0.0042$0.421M
68.1
Good value (68.1). Cheaper per task than its benchmark results predict. Plenty of data behind it.
60 Β±9
Chatty (60 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
157
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Hy 4
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
137.4
From 46 benchmarks. Plenty of data behind it.
β‰ˆ 123Β Β±12
Coding IQ 123.4, 95% interval 111–136. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.834cache read $0.042$2.5011M
56.3
Good value (56.3). Cheaper per task than its benchmark results predict. Plenty of data behind it.
61 Β±9
Chatty (61 Β±9). Generates more tokens than models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate.
Not measured
DeepSeek V4 Pro 0813
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
136.8
From 147 benchmarks. Plenty of data behind it.
117Β Β±7
Coding IQ 117.1, 95% interval 110–124. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$1.30cache read $0.10$2.601M
52.5
Fair value (52.5). Costs per task about what its benchmark results predict. Plenty of data behind it.
58 Β±9
Chatty (58 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
156
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Inkling 2 Small
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.
New
Released 5 Nov 2026
136.8
From 97 benchmarks. Plenty of data behind it.
117Β Β±8
Coding IQ 116.9, 95% interval 109–125. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.30cache read $0.03$1.001M
63.2
Good value (63.2). Cheaper per task than its benchmark results predict. Plenty of data behind it.
Not measuredNot measured
MiMo V2.6 Pro
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
136.7
From 133 benchmarks. Plenty of data behind it.
123Β Β±9
Coding IQ 123.4, 95% interval 115–132. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.43cache read $0.0036$0.871M
74.5
Good value (74.5). Cheaper per task than its benchmark results predict. Plenty of data behind it.
62 Β±9
Chatty (62 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
Not measured
DeepSeek V4 Flash Vision Exp
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
136.6
From 32 benchmarks. A fair amount of data behind it.
β‰ˆ 110Β Β±19
Coding IQ 109.6, 95% interval 90–129. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.44cache read $0.014$1.321M
63.0
Good value (63.0). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Hy4 preview
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
136.6
From 95 benchmarks. Plenty of data behind it.
118Β Β±16
Coding IQ 118.2, 95% interval 102–135. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.834cache read $0.042$2.5011M
56.4
Good value (56.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
57 Β±9
Chatty (57 Β±9). Generates more tokens than models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate.
Not measured
Claude Sonnet 5
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.
136.3
From 360 benchmarks. Plenty of data behind it.
115Β Β±5
Coding IQ 115.2, 95% interval 111–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$3.00$15.001M
37.4
Poor value (37.4). Costs more per task than its benchmark results predict. Plenty of data behind it.
57 Β±9
Chatty (57 Β±9). Generates more tokens than models with similar coding results. Measured at medium effort. Little data behind it, so treat it as a rough estimate.
156.9
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Gemini 3.1 Pro Preview
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
google/gemini-3.1-pro
136.0
From 595 benchmarks. Plenty of data behind it.
110Β Β±9
Coding IQ 110.0, 95% interval 102–119. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$2.00$12.001M
49.8
Fair value (49.8). Costs per task about what its benchmark results predict. Plenty of data behind it.
38 Β±12
Terse (38 Β±12). Generates fewer tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it.
156.3
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Gemini 3.5 Flash
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Newer version
Newer versionGemini 3.8 FlashWe compared release dates and version names in our catalogue.
135.9
From 393 benchmarks. Plenty of data behind it.
113Β Β±5
Coding IQ 113.3, 95% interval 109–118. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$1.50$9.001M
50.2
Fair value (50.2). Costs per task about what its benchmark results predict. Plenty of data behind it.
77 Β±6
Chatty (77 Β±6). Generates more tokens than models with similar coding results. Measured at default effort. A fair amount of data behind it.
155.3
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Qwen3.8 Flash
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
135.9
From 44 benchmarks. Plenty of data behind it.
Not enough data yet$0.113cache read $0.0141$0.3821M
65.8
Good value (65.8). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured

Run by DeepInfra

58 listings on this page where DeepInfra runs the model. Another provider sells these listings.

Select a score for details.

About these columns
llminfo IQ
llminfo IQ (capability index). How llminfo IQ is computed
Coding IQ
llminfo Coding IQ (coding capability, its own scale). How Coding IQ is computed
Value
Value for money. 50 means as expected for its results. Above 50, cheaper per task. How Value for money works
Yap
Tokens used per coding task vs. models with the same results. Includes agents that write code. Chattiness on maths doesn't predict it. How the Yap score works
ModelProviderllminfo IQCoding IQInput $/1MOutput $/1MMax Input TokensValueYapEpoch Capabilities Index
Qwen4 Max
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
New
Released 11 Nov 2026
πŸ‡ΊπŸ‡Έ US
Β NanoGPT
144.9
From 40 benchmarks. Plenty of data behind it.
β‰ˆ 124Β Β±12
Coding IQ 124.3, 95% interval 112–137. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$1.4837cache read $0.0746$4.452Not published by the provider
61.7
Good value (61.7). Cheaper per task than its benchmark results predict. Plenty of data behind it.
Not measuredNot measured
Qwen4 Max
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
New
Released 11 Nov 2026
qwen/qwen4-max:thinking Β· thinking Β· πŸ‡ΊπŸ‡Έ US
Β NanoGPT
144.9
From 40 benchmarks. Plenty of data behind it.
β‰ˆ 124Β Β±12
Coding IQ 124.3, 95% interval 112–137. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$1.4837cache read $0.0746$4.452Not published by the provider
61.7
Good value (61.7). Cheaper per task than its benchmark results predict. Plenty of data behind it.
Not measuredNot measured
Kimi K3.1 Pro
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.
Β Hugging Face
142.7
From 161 benchmarks. Plenty of data behind it.
131Β Β±10
Coding IQ 131.3, 95% interval 121–141. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$1.90$9.501M
49.2
Fair value (49.2). Costs per task about what its benchmark results predict. Plenty of data behind it.
51 Β±9
Typical (51 Β±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
161.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Kimi K3.1 Pro
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
Β OpenRouter
142.7
From 161 benchmarks. Plenty of data behind it.
131Β Β±10
Coding IQ 131.3, 95% interval 121–141. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$1.90cache read $0.19$9.501M
49.2
Fair value (49.2). Costs per task about what its benchmark results predict. Plenty of data behind it.
51 Β±9
Typical (51 Β±9). Generates about as many tokens as models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
161.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4.1 Flash 1017
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.
Β Eden AI
139.7
From 97 benchmarks. Plenty of data behind it.
127Β Β±8
Coding IQ 126.8, 95% interval 119–135. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.20cache read $0.006$0.601M
71.4
Good value (71.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
55 Β±9
Typical (55 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.
159.4
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4.1 Flash 1017
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.
Β Hugging Face
139.7
From 97 benchmarks. Plenty of data behind it.
127Β Β±8
Coding IQ 126.8, 95% interval 119–135. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.098$0.2941M
71.4
Good value (71.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
55 Β±9
Typical (55 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.
159.4
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4.1 Flash 1017
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.
Β Llmgateway
139.7
From 97 benchmarks. Plenty of data behind it.
127Β Β±8
Coding IQ 126.8, 95% interval 119–135. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.20cache read $0.006$0.60Not published by the provider
71.4
Good value (71.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
55 Β±9
Typical (55 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.
159.4
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4.1 Flash 1017
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.
latest
Β Llmgateway
139.7
From 97 benchmarks. Plenty of data behind it.
127Β Β±8
Coding IQ 126.8, 95% interval 119–135. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.20cache read $0.006$0.60Not published by the provider
71.4
Good value (71.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
55 Β±9
Typical (55 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.
159.4
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4.1 Flash 1017
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 Β· πŸ‡ΊπŸ‡Έ US
Β NanoGPT
139.7
From 97 benchmarks. Plenty of data behind it.
127Β Β±8
Coding IQ 126.8, 95% interval 119–135. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.147cache read $0.0044$0.441Not published by the provider
71.4
Good value (71.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
55 Β±9
Typical (55 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.
159.4
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4.1 Flash 1017
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.
deepseek/deepseek-v4.1-flash-1017:thinking Β· thinking Β· fp8 Β· πŸ‡ΊπŸ‡Έ US
Β NanoGPT
139.7
From 97 benchmarks. Plenty of data behind it.
127Β Β±8
Coding IQ 126.8, 95% interval 119–135. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.147cache read $0.0044$0.441Not published by the provider
71.4
Good value (71.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
55 Β±9
Typical (55 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.
159.4
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4.1 Flash 1017
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
Β OpenRouter
139.7
From 97 benchmarks. Plenty of data behind it.
127Β Β±8
Coding IQ 126.8, 95% interval 119–135. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.098cache read $0.0029$0.2941M
71.4
Good value (71.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
55 Β±9
Typical (55 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.
159.4
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4.1 Flash 1017
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
Β OpenRouter
139.7
From 97 benchmarks. Plenty of data behind it.
127Β Β±8
Coding IQ 126.8, 95% interval 119–135. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.0784cache read $0.0024$0.23521M
71.4
Good value (71.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
55 Β±9
Typical (55 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.
159.4
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4.1 Flash 1017
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
Β OpenRouter
139.7
From 97 benchmarks. Plenty of data behind it.
127Β Β±8
Coding IQ 126.8, 95% interval 119–135. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.147cache read $0.0044$0.4411M
71.4
Good value (71.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
55 Β±9
Typical (55 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.
159.4
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Kimi K3
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.
Β Hugging Face
139.2
From 417 benchmarks. Plenty of data behind it.
124Β Β±7
Coding IQ 123.8, 95% interval 117–131. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$2.85$14.251M
60.2
Good value (60.2). Cheaper per task than its benchmark results predict. Plenty of data behind it.
55 Β±9
Typical (55 Β±9). Generates about as many tokens as models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate.
158.6
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Kimi K3
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.
mxfp4 Β· πŸ‡ΊπŸ‡Έ US
Β NanoGPT
139.2
From 417 benchmarks. Plenty of data behind it.
124Β Β±7
Coding IQ 123.8, 95% interval 117–131. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$2.9925cache read $0.2993$14.963Not published by the provider
60.2
Good value (60.2). Cheaper per task than its benchmark results predict. Plenty of data behind it.
55 Β±9
Typical (55 Β±9). Generates about as many tokens as models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate.
158.6
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Kimi K3
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-k3:batch Β· batch+bf16 Β· bf16
Β OpenRouter
139.2
From 417 benchmarks. Plenty of data behind it.
124Β Β±7
Coding IQ 123.8, 95% interval 117–131. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$2.28cache read $0.228$11.401M
60.2
Good value (60.2). Cheaper per task than its benchmark results predict. Plenty of data behind it.
55 Β±9
Typical (55 Β±9). Generates about as many tokens as models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate.
158.6
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Kimi K3
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.
mxfp4
Β OpenRouter
139.2
From 417 benchmarks. Plenty of data behind it.
124Β Β±7
Coding IQ 123.8, 95% interval 117–131. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$2.85cache read $0.285$14.251M
60.2
Good value (60.2). Cheaper per task than its benchmark results predict. Plenty of data behind it.
55 Β±9
Typical (55 Β±9). Generates about as many tokens as models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate.
158.6
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Qwen3.8 Max
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
Newer version
Newer versionQwen4 MaxWe compared release dates and version names in our catalogue.
Qwen/Qwen3.8-2.4T-A95B
Β Hugging Face
138.7
From 312 benchmarks. Plenty of data behind it.
117Β Β±5
Coding IQ 117.2, 95% interval 112–122. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$2.00$6.00262K
49.2
Fair value (49.2). Costs per task about what its benchmark results predict. Plenty of data behind it.
62 Β±9
Chatty (62 Β±9). Generates more tokens than models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate.
157.6
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Qwen3.8 Max
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
Newer version
Newer versionQwen4 MaxWe compared release dates and version names in our catalogue.
Qwen/Qwen3.8-2.4T-A95B
Β Llmgateway
138.7
From 312 benchmarks. Plenty of data behind it.
117Β Β±5
Coding IQ 117.2, 95% interval 112–122. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$2.00cache read $0.20$6.00Not published by the provider
49.2
Fair value (49.2). Costs per task about what its benchmark results predict. Plenty of data behind it.
62 Β±9
Chatty (62 Β±9). Generates more tokens than models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate.
157.6
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GLM 5.3
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.
Β Hugging Face
138.2
From 318 benchmarks. Plenty of data behind it.
121Β Β±9
Coding IQ 121.0, 95% interval 112–130. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.5625$2.501M
57.7
Good value (57.7). Cheaper per task than its benchmark results predict. Plenty of data behind it.
65 Β±10
Chatty (65 Β±10). Generates more tokens than models with similar coding results. Measured at max effort. A fair amount of data behind it.
157.1
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GLM 5.3
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.
Β Llmgateway
138.2
From 318 benchmarks. Plenty of data behind it.
121Β Β±9
Coding IQ 121.0, 95% interval 112–130. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$1.20cache read $0.20$4.00Not published by the provider
57.7
Good value (57.7). Cheaper per task than its benchmark results predict. Plenty of data behind it.
65 Β±10
Chatty (65 Β±10). Generates more tokens than models with similar coding results. Measured at max effort. A fair amount of data behind it.
157.1
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GLM 5.3
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 Β· πŸ‡ΊπŸ‡Έ US
Β NanoGPT
138.2
From 318 benchmarks. Plenty of data behind it.
121Β Β±9
Coding IQ 121.0, 95% interval 112–130. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.5906cache read $0.1313$2.625Not published by the provider
57.7
Good value (57.7). Cheaper per task than its benchmark results predict. Plenty of data behind it.
65 Β±10
Chatty (65 Β±10). Generates more tokens than models with similar coding results. Measured at max effort. A fair amount of data behind it.
157.1
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GLM 5.3
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-5.3:thinking Β· thinking Β· fp4 Β· πŸ‡ΊπŸ‡Έ US
Β NanoGPT
138.2
From 318 benchmarks. Plenty of data behind it.
121Β Β±9
Coding IQ 121.0, 95% interval 112–130. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.5906cache read $0.1313$2.625Not published by the provider
57.7
Good value (57.7). Cheaper per task than its benchmark results predict. Plenty of data behind it.
65 Β±10
Chatty (65 Β±10). Generates more tokens than models with similar coding results. Measured at max effort. A fair amount of data behind it.
157.1
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GLM 5.3
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-5.3:batch Β· batch+fp4 Β· fp4
Β OpenRouter
138.2
From 318 benchmarks. Plenty of data behind it.
121Β Β±9
Coding IQ 121.0, 95% interval 112–130. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.45cache read $0.10$2.001M
57.7
Good value (57.7). Cheaper per task than its benchmark results predict. Plenty of data behind it.
65 Β±10
Chatty (65 Β±10). Generates more tokens than models with similar coding results. Measured at max effort. A fair amount of data behind it.
157.1
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GLM 5.3
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
Β OpenRouter
138.2
From 318 benchmarks. Plenty of data behind it.
121Β Β±9
Coding IQ 121.0, 95% interval 112–130. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.5625cache read $0.125$2.501M
57.7
Good value (57.7). Cheaper per task than its benchmark results predict. Plenty of data behind it.
65 Β±10
Chatty (65 Β±10). Generates more tokens than models with similar coding results. Measured at max effort. A fair amount of data behind it.
157.1
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GLM 5.3
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
Β OpenRouter
138.2
From 318 benchmarks. Plenty of data behind it.
121Β Β±9
Coding IQ 121.0, 95% interval 112–130. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.45cache read $0.10$2.001M
57.7
Good value (57.7). Cheaper per task than its benchmark results predict. Plenty of data behind it.
65 Β±10
Chatty (65 Β±10). Generates more tokens than models with similar coding results. Measured at max effort. A fair amount of data behind it.
157.1
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4.1 Flash
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
Β Eden AI
137.7
From 179 benchmarks. Plenty of data behind it.
125Β Β±16
Coding IQ 124.7, 95% interval 109–140. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.20cache read $0.006$0.601M
68.1
Good value (68.1). Cheaper per task than its benchmark results predict. Plenty of data behind it.
60 Β±9
Chatty (60 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
157
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4.1 Flash
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
Β Hugging Face
137.7
From 179 benchmarks. Plenty of data behind it.
125Β Β±16
Coding IQ 124.7, 95% interval 109–140. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.14$0.421M
68.1
Good value (68.1). Cheaper per task than its benchmark results predict. Plenty of data behind it.
60 Β±9
Chatty (60 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
157
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4.1 Flash
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
Β Llmgateway
137.7
From 179 benchmarks. Plenty of data behind it.
125Β Β±16
Coding IQ 124.7, 95% interval 109–140. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.20cache read $0.006$0.60Not published by the provider
68.1
Good value (68.1). Cheaper per task than its benchmark results predict. Plenty of data behind it.
60 Β±9
Chatty (60 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
157
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4.1 Flash
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
fp8 Β· πŸ‡ΊπŸ‡Έ US
Β NanoGPT
137.7
From 179 benchmarks. Plenty of data behind it.
125Β Β±16
Coding IQ 124.7, 95% interval 109–140. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.147cache read $0.0044$0.441Not published by the provider
68.1
Good value (68.1). Cheaper per task than its benchmark results predict. Plenty of data behind it.
60 Β±9
Chatty (60 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
157
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4.1 Flash
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
deepseek/deepseek-v4.1-flash:thinking Β· thinking Β· fp8 Β· πŸ‡ΊπŸ‡Έ US
Β NanoGPT
137.7
From 179 benchmarks. Plenty of data behind it.
125Β Β±16
Coding IQ 124.7, 95% interval 109–140. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.147cache read $0.0044$0.441Not published by the provider
68.1
Good value (68.1). Cheaper per task than its benchmark results predict. Plenty of data behind it.
60 Β±9
Chatty (60 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
157
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Hy 4
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
Β Hugging Face
137.4
From 46 benchmarks. Plenty of data behind it.
β‰ˆ 123Β Β±12
Coding IQ 123.4, 95% interval 111–136. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.834$2.5011M
56.3
Good value (56.3). Cheaper per task than its benchmark results predict. Plenty of data behind it.
61 Β±9
Chatty (61 Β±9). Generates more tokens than models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate.
Not measured
Hy 4
πŸ‡¨πŸ‡³
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
fp4
Β OpenRouter
137.4
From 46 benchmarks. Plenty of data behind it.
β‰ˆ 123Β Β±12
Coding IQ 123.4, 95% interval 111–136. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.834cache read $0.042$2.5011M
56.3
Good value (56.3). Cheaper per task than its benchmark results predict. Plenty of data behind it.
61 Β±9
Chatty (61 Β±9). Generates more tokens than models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate.
Not measured
DeepSeek V4 Pro 0813
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.
Β Eden AI
136.8
From 147 benchmarks. Plenty of data behind it.
117Β Β±7
Coding IQ 117.1, 95% interval 110–124. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$1.30cache read $0.10$2.601M
52.5
Fair value (52.5). Costs per task about what its benchmark results predict. Plenty of data behind it.
58 Β±9
Chatty (58 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
156
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4 Pro 0813
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.
Β Hugging Face
136.8
From 147 benchmarks. Plenty of data behind it.
117Β Β±7
Coding IQ 117.1, 95% interval 110–124. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$1.30$2.601M
52.5
Fair value (52.5). Costs per task about what its benchmark results predict. Plenty of data behind it.
58 Β±9
Chatty (58 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
156
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4 Pro 0813
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 Β· πŸ‡ΊπŸ‡Έ US
Β NanoGPT
136.8
From 147 benchmarks. Plenty of data behind it.
117Β Β±7
Coding IQ 117.1, 95% interval 110–124. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$1.365cache read $0.105$2.73Not published by the provider
52.5
Fair value (52.5). Costs per task about what its benchmark results predict. Plenty of data behind it.
58 Β±9
Chatty (58 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
156
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4 Pro 0813
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.
deepseek/deepseek-v4-pro-0813:thinking Β· thinking Β· fp8 Β· πŸ‡ΊπŸ‡Έ US
Β NanoGPT
136.8
From 147 benchmarks. Plenty of data behind it.
117Β Β±7
Coding IQ 117.1, 95% interval 110–124. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$1.365cache read $0.105$2.73Not published by the provider
52.5
Fair value (52.5). Costs per task about what its benchmark results predict. Plenty of data behind it.
58 Β±9
Chatty (58 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
156
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4 Pro 0813
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
Β OpenRouter
136.8
From 147 benchmarks. Plenty of data behind it.
117Β Β±7
Coding IQ 117.1, 95% interval 110–124. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$1.30cache read $0.10$2.601M
52.5
Fair value (52.5). Costs per task about what its benchmark results predict. Plenty of data behind it.
58 Β±9
Chatty (58 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
156
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4 Pro 0813
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
Β OpenRouter
136.8
From 147 benchmarks. Plenty of data behind it.
117Β Β±7
Coding IQ 117.1, 95% interval 110–124. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$1.04cache read $0.08$2.081M
52.5
Fair value (52.5). Costs per task about what its benchmark results predict. Plenty of data behind it.
58 Β±9
Chatty (58 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
156
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4 Pro 0813
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
Β OpenRouter
136.8
From 147 benchmarks. Plenty of data behind it.
117Β Β±7
Coding IQ 117.1, 95% interval 110–124. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$1.95cache read $0.15$3.901M
52.5
Fair value (52.5). Costs per task about what its benchmark results predict. Plenty of data behind it.
58 Β±9
Chatty (58 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
156
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Inkling 2 Small
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.
New
Released 5 Nov 2026
Β Eden AI
136.8
From 97 benchmarks. Plenty of data behind it.
117Β Β±8
Coding IQ 116.9, 95% interval 109–125. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.30cache read $0.03$1.001M
63.2
Good value (63.2). Cheaper per task than its benchmark results predict. Plenty of data behind it.
Not measuredNot measured
Inkling 2 Small
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.
New
Released 5 Nov 2026
Β Hugging Face
136.8
From 97 benchmarks. Plenty of data behind it.
117Β Β±8
Coding IQ 116.9, 95% interval 109–125. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.285$0.951M
63.2
Good value (63.2). Cheaper per task than its benchmark results predict. Plenty of data behind it.
Not measuredNot measured
Inkling 2 Small
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.
New
Released 5 Nov 2026
fp4
Β OpenRouter
136.8
From 97 benchmarks. Plenty of data behind it.
117Β Β±8
Coding IQ 116.9, 95% interval 109–125. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.285cache read $0.0285$0.951M
63.2
Good value (63.2). Cheaper per task than its benchmark results predict. Plenty of data behind it.
Not measuredNot measured
MiMo V2.6 Pro
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.
Β Llmgateway
136.7
From 133 benchmarks. Plenty of data behind it.
123Β Β±9
Coding IQ 123.4, 95% interval 115–132. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.435cache read $0.0036$0.87Not published by the provider
74.5
Good value (74.5). Cheaper per task than its benchmark results predict. Plenty of data behind it.
62 Β±9
Chatty (62 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
Not measured
MiMo V2.6 Pro
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 Β· πŸ‡ΊπŸ‡Έ US
Β NanoGPT
136.7
From 133 benchmarks. Plenty of data behind it.
123Β Β±9
Coding IQ 123.4, 95% interval 115–132. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.4064cache read $0.0034$0.8222Not published by the provider
74.5
Good value (74.5). Cheaper per task than its benchmark results predict. Plenty of data behind it.
62 Β±9
Chatty (62 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
Not measured
MiMo V2.6 Pro
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
Β OpenRouter
136.7
From 133 benchmarks. Plenty of data behind it.
123Β Β±9
Coding IQ 123.4, 95% interval 115–132. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.387cache read $0.0032$0.7831M
74.5
Good value (74.5). Cheaper per task than its benchmark results predict. Plenty of data behind it.
62 Β±9
Chatty (62 Β±9). Generates more tokens than models with similar coding results. Measured at default effort. Little data behind it, so treat it as a rough estimate.
Not measured
DeepSeek V4 Flash Vision Exp
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.
Β Hugging Face
136.6
From 32 benchmarks. A fair amount of data behind it.
β‰ˆ 110Β Β±19
Coding IQ 109.6, 95% interval 90–129. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.2156$0.64681M
63.0
Good value (63.0). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
DeepSeek V4 Flash Vision Exp
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.
Β Llmgateway
136.6
From 32 benchmarks. A fair amount of data behind it.
β‰ˆ 110Β Β±19
Coding IQ 109.6, 95% interval 90–129. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.44$1.32Not published by the provider
63.0
Good value (63.0). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
DeepSeek V4 Flash Vision Exp
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 Β· πŸ‡ΊπŸ‡Έ US
Β NanoGPT
136.6
From 32 benchmarks. A fair amount of data behind it.
β‰ˆ 110Β Β±19
Coding IQ 109.6, 95% interval 90–129. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.2264cache read $0.0072$0.6791Not published by the provider
63.0
Good value (63.0). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
DeepSeek V4 Flash Vision Exp
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
Β OpenRouter
136.6
From 32 benchmarks. A fair amount of data behind it.
β‰ˆ 110Β Β±19
Coding IQ 109.6, 95% interval 90–129. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.2156cache read $0.0069$0.64681M
63.0
Good value (63.0). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
DeepSeek V4 Flash Vision Exp
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
Β OpenRouter
136.6
From 32 benchmarks. A fair amount of data behind it.
β‰ˆ 110Β Β±19
Coding IQ 109.6, 95% interval 90–129. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.1725cache read $0.0055$0.51741M
63.0
Good value (63.0). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
DeepSeek V4 Flash Vision Exp
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
Β OpenRouter
136.6
From 32 benchmarks. A fair amount of data behind it.
β‰ˆ 110Β Β±19
Coding IQ 109.6, 95% interval 90–129. Little data behind it, so treat it as a rough estimate. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.3234cache read $0.0103$0.97021M
63.0
Good value (63.0). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measuredNot measured
Hy4 preview
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.
Β Eden AI
136.6
From 95 benchmarks. Plenty of data behind it.
118Β Β±16
Coding IQ 118.2, 95% interval 102–135. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.834cache read $0.042$2.5011M
56.4
Good value (56.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
57 Β±9
Chatty (57 Β±9). Generates more tokens than models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate.
Not measured
Hy4 preview
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.
Β Hugging Face
136.6
From 95 benchmarks. Plenty of data behind it.
118Β Β±16
Coding IQ 118.2, 95% interval 102–135. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.834$2.5011M
56.4
Good value (56.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
57 Β±9
Chatty (57 Β±9). Generates more tokens than models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate.
Not measured
Hy4 preview
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.
Β Llmgateway
136.6
From 95 benchmarks. Plenty of data behind it.
118Β Β±16
Coding IQ 118.2, 95% interval 102–135. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.834cache read $0.042$2.501Not published by the provider
56.4
Good value (56.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
57 Β±9
Chatty (57 Β±9). Generates more tokens than models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate.
Not measured
Hy4 preview
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 Β· πŸ‡ΊπŸ‡Έ US
Β NanoGPT
136.6
From 95 benchmarks. Plenty of data behind it.
118Β Β±16
Coding IQ 118.2, 95% interval 102–135. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.8757cache read $0.0441$2.6261Not published by the provider
56.4
Good value (56.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
57 Β±9
Chatty (57 Β±9). Generates more tokens than models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate.
Not measured
Hy4 preview
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
Β OpenRouter
136.6
From 95 benchmarks. Plenty of data behind it.
118Β Β±16
Coding IQ 118.2, 95% interval 102–135. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.834cache read $0.042$2.5011M
56.4
Good value (56.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
57 Β±9
Chatty (57 Β±9). Generates more tokens than models with similar coding results. Measured at max effort. Little data behind it, so treat it as a rough estimate.
Not measured
Ling 3.1 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 versionLing 3.2 FlashWe compared the version names in our catalogue.
fp4 Β· πŸ‡ΊπŸ‡Έ US
Β NanoGPT
135.2
From 16 benchmarks. A fair amount of data behind it.
Not enough data yet$0.315cache read $0.063$0.945Not published by the provider
69.6
Good value (69.6). Cheaper 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