Skip to content

Listings

All listings: Neuralwatt

Every listing for Neuralwatt's models, including each model's other versions and routes. Back to Neuralwatt

Sold by Neuralwatt

24 listings sold by Neuralwatt.

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
Kimi K3.1 Pro
🇨🇳
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.
131 ±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 ±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.
161.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Kimi K3.1 Pro
🇨🇳
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.
kimi-k3.1-pro-flex · flex
142.7
From 161 benchmarks. Plenty of data behind it.
131 ±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.
$1.30cache read $0.13$6.501M
49.2
Fair value (49.2). Costs per task about what 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.
161.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GLM 5.3
🇨🇳
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.
138.2
From 318 benchmarks. Plenty of data behind it.
121 ±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.
$1.45cache read $0.145$4.501M
57.7
Good value (57.7). Cheaper per task than its benchmark results predict. Plenty of data behind it.
65 ±10
Chatty (65 ±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
🇨🇳
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.
glm-5.3-flex · flex
138.2
From 318 benchmarks. Plenty of data behind it.
121 ±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.
$0.9425cache read $0.0943$2.9251M
57.7
Good value (57.7). Cheaper per task than its benchmark results predict. Plenty of data behind it.
65 ±10
Chatty (65 ±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
🇨🇳
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
137.7
From 179 benchmarks. Plenty of data behind it.
125 ±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.
$0.15cache read $0.015$0.601M
68.1
Good value (68.1). Cheaper per task than its benchmark results predict. Plenty of data behind it.
60 ±9
Chatty (60 ±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
🇨🇳
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
deepseek-v4.1-flash-flex · flex
137.7
From 179 benchmarks. Plenty of data behind it.
125 ±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.
$0.0975cache read $0.0098$0.391M
68.1
Good value (68.1). Cheaper per task than its benchmark results predict. Plenty of data behind it.
60 ±9
Chatty (60 ±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
MiMo V2.6 Pro
🇨🇳
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.
136.7
From 133 benchmarks. Plenty of data behind it.
123 ±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$0.871M
74.5
Good value (74.5). Cheaper per task than its benchmark results predict. Plenty of data behind it.
62 ±9
Chatty (62 ±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
🇨🇳
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.
mimo-v2.6-pro-flex · flex
136.7
From 133 benchmarks. Plenty of data behind it.
123 ±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.
$0.2828cache read $0.0023$0.56551M
74.5
Good value (74.5). Cheaper per task than its benchmark results predict. Plenty of data behind it.
62 ±9
Chatty (62 ±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
GLM 5.3 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.
fp8
133.6
From 266 benchmarks. Plenty of data behind it.
113 ±8
Coding IQ 112.8, 95% interval 105–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.15cache read $0.03$0.501M
65.3
Good value (65.3). Cheaper per task than its benchmark results predict. Plenty of data behind it.
68 ±11
Chatty (68 ±11). Generates more tokens than models with similar coding results. Measured at max effort. A fair amount of data behind it.
153.8
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GLM 5.3 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.
glm-5.3-flash-flex · flex · fp8
133.6
From 266 benchmarks. Plenty of data behind it.
113 ±8
Coding IQ 112.8, 95% interval 105–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.0975cache read $0.0195$0.3251M
65.3
Good value (65.3). Cheaper per task than its benchmark results predict. Plenty of data behind it.
68 ±11
Chatty (68 ±11). Generates more tokens than models with similar coding results. Measured at max effort. A fair amount of data behind it.
153.8
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Qwen 3.8 27B
🇨🇳
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
129.4
From 251 benchmarks. Plenty of data behind it.
111 ±7
Coding IQ 111.4, 95% interval 104–119. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.45cache read $0.25$3.20262K
39.7
Poor value (39.7). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measured
152.6
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Qwen 3.8 27B
🇨🇳
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
qwen-3.8-27b-flex · flex
129.4
From 251 benchmarks. Plenty of data behind it.
111 ±7
Coding IQ 111.4, 95% interval 104–119. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.2925cache read $0.1625$2.08262K
39.7
Poor value (39.7). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measured
152.6
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Gemma 4 31B IT
🇺🇸
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.
gemma-4-31b · fp4
128.3
From 333 benchmarks. Plenty of data behind it.
102 ±7
Coding IQ 101.7, 95% interval 94–109. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.144cache read $0.0144$0.42262K
61.9
Good value (61.9). Cheaper per task than its benchmark results predict. Plenty 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.
144.1
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Neuralwatt 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.
nw-flash
Not enough data yetNot enough data yet$0.15cache read $0.015$0.601MNot enough data yetNot measuredNot measured
Neuralwatt 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.
nw-flash-flex · flex
Not enough data yetNot enough data yet$0.0975cache read $0.0098$0.391MNot enough data yetNot measuredNot measured
DeepSeek V4.1 Flash (Speed)
🇨🇳
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.
Not enough data yetNot enough data yet$0.15cache read $0.015$0.601MNot enough data yetNot measuredNot measured
Mammoth 1
🇺🇸
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.
Doesn’t apply to this modelDoesn’t apply to this model$0.18cache read $0.018Doesn’t apply to this model33KDoesn’t apply to this modelDoesn’t apply to this modelDoesn’t apply to this model
Qwen3.6 35B
🇨🇳
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.
Not enough data yetNot enough data yet$0.29cache read $0.029$1.15262KNot enough data yetNot measuredNot measured
Qwen3.6 35B
🇨🇳
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.
qwen3.6-35b-fast · fast
Not enough data yetNot enough data yet$0.29cache read $0.029$1.15262KNot enough data yetNot measuredNot measured
Qwen3.6 35B
🇨🇳
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.
qwen3.6-35b-flex · flex
Not enough data yetNot enough data yet$0.1885cache read $0.0189$0.7475262KNot enough data yetNot measuredNot measured
Neuralwatt Small
🇨🇳
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.
nw-small
Not enough data yetNot enough data yet$0.45cache read $0.25$3.20262KNot enough data yetNot measuredNot measured
Neuralwatt Small
🇨🇳
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.
nw-small-flex · flex
Not enough data yetNot enough data yet$0.2925cache read $0.1625$2.08262KNot enough data yetNot measuredNot measured
Neuralwatt Large
🇨🇳
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.
nw-large
Not enough data yetNot enough data yet$2.00cache read $0.20$10.001MNot enough data yetNot measuredNot measured
Neuralwatt Large
🇨🇳
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.
nw-large-flex · flex
Not enough data yetNot enough data yet$1.30cache read $0.13$6.501MNot enough data yetNot measuredNot measured

Run by Neuralwatt

19 listings where Neuralwatt 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
Kimi K3.1 Pro
🇨🇳
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.
🇺🇸 US
 NanoGPT
142.7
From 161 benchmarks. Plenty of data behind it.
131 ±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.
$2.10cache read $0.21$10.50Not published by the provider
49.2
Fair value (49.2). Costs per task about what 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.
161.2
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4.1 Flash 1017
🇨🇳
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.
🇺🇸 US
 NanoGPT
139.7
From 97 benchmarks. Plenty of data behind it.
127 ±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.
$0.1575cache read $0.0158$0.63Not published by the provider
71.4
Good value (71.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
55 ±9
Typical (55 ±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
🇨🇳
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 · thinking · 🇺🇸 US
 NanoGPT
139.7
From 97 benchmarks. Plenty of data behind it.
127 ±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.
$0.1575cache read $0.0158$0.63Not published by the provider
71.4
Good value (71.4). Cheaper per task than its benchmark results predict. Plenty of data behind it.
55 ±9
Typical (55 ±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
GLM 5.3
🇨🇳
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.
🇺🇸 US
 NanoGPT
138.2
From 318 benchmarks. Plenty of data behind it.
121 ±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.
$1.5225cache read $0.1523$4.725Not published by the provider
57.7
Good value (57.7). Cheaper per task than its benchmark results predict. Plenty of data behind it.
65 ±10
Chatty (65 ±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
🇨🇳
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 · thinking · 🇺🇸 US
 NanoGPT
138.2
From 318 benchmarks. Plenty of data behind it.
121 ±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.
$1.5225cache read $0.1523$4.725Not published by the provider
57.7
Good value (57.7). Cheaper per task than its benchmark results predict. Plenty of data behind it.
65 ±10
Chatty (65 ±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
MiMo V2.6 Pro
🇨🇳
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.
🇺🇸 US
 NanoGPT
136.7
From 133 benchmarks. Plenty of data behind it.
123 ±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.
$0.4568cache read $0.0038$0.9135Not published by the provider
74.5
Good value (74.5). Cheaper per task than its benchmark results predict. Plenty of data behind it.
62 ±9
Chatty (62 ±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
🇨🇳
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.
flex · 🇺🇸 US
 NanoGPT
136.7
From 133 benchmarks. Plenty of data behind it.
123 ±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.
$0.2969cache read $0.0025$0.5938Not published by the provider
74.5
Good value (74.5). Cheaper per task than its benchmark results predict. Plenty of data behind it.
62 ±9
Chatty (62 ±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
GLM 5.3 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.
fp8 · 🇺🇸 US
 NanoGPT
133.6
From 266 benchmarks. Plenty of data behind it.
113 ±8
Coding IQ 112.8, 95% interval 105–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.1575cache read $0.0315$0.525Not published by the provider
65.3
Good value (65.3). Cheaper per task than its benchmark results predict. Plenty of data behind it.
68 ±11
Chatty (68 ±11). Generates more tokens than models with similar coding results. Measured at max effort. A fair amount of data behind it.
153.8
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
GLM 5.3 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.
flex · fp8 · 🇺🇸 US
 NanoGPT
133.6
From 266 benchmarks. Plenty of data behind it.
113 ±8
Coding IQ 112.8, 95% interval 105–120. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.1024cache read $0.0205$0.3413Not published by the provider
65.3
Good value (65.3). Cheaper per task than its benchmark results predict. Plenty of data behind it.
68 ±11
Chatty (68 ±11). Generates more tokens than models with similar coding results. Measured at max effort. A fair amount of data behind it.
153.8
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Qwen 3.8 27B
🇨🇳
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
 NanoGPT
129.4
From 251 benchmarks. Plenty of data behind it.
111 ±7
Coding IQ 111.4, 95% interval 104–119. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.4725cache read $0.2625$3.36Not published by the provider
39.7
Poor value (39.7). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measured
152.6
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Qwen 3.8 27B
🇨🇳
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
qwen/qwen3.8-27b:thinking · thinking · fp8 · 🇺🇸 US
 NanoGPT
129.4
From 251 benchmarks. Plenty of data behind it.
111 ±7
Coding IQ 111.4, 95% interval 104–119. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.4725cache read $0.2625$3.36Not published by the provider
39.7
Poor value (39.7). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measured
152.6
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
DeepSeek V4 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 versionDeepSeek V4.1 Flash 1017We compared release dates and version names in our catalogue.
deepseek/deepseek-v4-flash-latest · latest · 🇺🇸 US
 NanoGPT
129.1
From 229 benchmarks. Plenty of data behind it.
108 ±12
Coding IQ 107.8, 95% interval 95–120. 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
74.6
Good value (74.6). Cheaper per task than its benchmark results predict. Plenty of data behind it.
44 ±9
Terse (44 ±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.
146.9
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Gemma 4 31B IT
🇺🇸
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
 NanoGPT
128.3
From 333 benchmarks. Plenty of data behind it.
102 ±7
Coding IQ 101.7, 95% interval 94–109. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.1512cache read $0.0151$0.441Not published by the provider
61.9
Good value (61.9). Cheaper per task than its benchmark results predict. Plenty 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.
144.1
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Gemma 4 31B IT
🇺🇸
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.
google/gemma-4-31b-it:thinking · thinking · fp4 · 🇺🇸 US
 NanoGPT
128.3
From 333 benchmarks. Plenty of data behind it.
102 ±7
Coding IQ 101.7, 95% interval 94–109. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.1512cache read $0.0151$0.441Not published by the provider
61.9
Good value (61.9). Cheaper per task than its benchmark results predict. Plenty 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.
144.1
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Qwen3.6 35B A3B
🇨🇳
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
🇺🇸 US
 NanoGPT
119.8
From 160 benchmarks. Plenty of data behind it.
95 ±14
Coding IQ 94.8, 95% interval 81–109. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.3045cache read $0.0305$1.2075Not published by the provider
38.5
Poor value (38.5). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measured
144.1
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Qwen3.6 35B A3B
🇨🇳
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
qwen/qwen3.6-35b-a3b:thinking · thinking · 🇺🇸 US
 NanoGPT
119.8
From 160 benchmarks. Plenty of data behind it.
95 ±14
Coding IQ 94.8, 95% interval 81–109. Plenty of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.3045cache read $0.0305$1.2075Not published by the provider
38.5
Poor value (38.5). Costs more per task than its benchmark results predict. Plenty of data behind it.
Not measured
144.1
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Hermes High
🇺🇸
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Router
This is a router, not a fixed model. The provider chooses which model answers, and that choice can change.
🇺🇸 US
 NanoGPTNot enough data yetNot enough data yet$1.5225cache read $0.1523$4.725Not published by the providerNot enough data yetNot measuredNot measured
Hermes Low
🇺🇸
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Router
This is a router, not a fixed model. The provider chooses which model answers, and that choice can change.
🇺🇸 US
 NanoGPTNot enough data yetNot enough data yet$1.5225cache read $0.1523$4.725Not published by the providerNot enough data yetNot measuredNot measured
Hermes Medium
🇺🇸
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Router
This is a router, not a fixed model. The provider chooses which model answers, and that choice can change.
🇺🇸 US
 NanoGPTNot enough data yetNot enough data yet$1.5225cache read $0.1523$4.725Not published by the providerNot enough data yetNot measuredNot measured

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

How this data works