Skip to content

Listings

All listings: NanoGPT

Every listing for the models on page 11 of 40 of NanoGPT, including each model's other versions and routes. Back to NanoGPT

Sold by NanoGPT

43 listings on this page sold by NanoGPT.

Select a score for details.

About these columns
llminfo IQ
llminfo IQ (capability index). How llminfo IQ is computed
Coding IQ
llminfo Coding IQ (coding capability, its own scale). How Coding IQ is computed
Value
Value for money. 50 means as expected for its results. Above 50, cheaper per task. How Value for money works
Modelllminfo IQCoding IQInput $/1MOutput $/1MMax Input TokensValueEpoch Capabilities Index
Phi 4 Multimodal Instruct
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.
103.4
From 48 benchmarks. Plenty of data behind it.
Not enough data yet$0.07cache read $0.035$0.11128K
59.1
Good value (59.1). Cheaper per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measured
L3.3 Ms Nevoria 70b
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
102.7
From 6 benchmarks. A fair amount of data behind it.
Not enough data yet$0.493cache read $0.2465$0.49333KNot enough data yetNot measured
Qwen2 72B Dracarys
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
abacusai/Dracarys-72B-Instruct
102.3
From 6 benchmarks. A fair amount of data behind it.
Not enough data yet$0.493cache read $0.2465$0.49333KNot enough data yetNot measured
Veo 3.1 Fast
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
veo3-1-fast-video
β‰ˆ 102
From 10 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yetβ€”β€”Not published by the providerNot enough data yetNot measured
Veo 3.1 Fast
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
veo3-1-fast-extend Β· extend
β‰ˆ 102
From 10 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yetβ€”β€”Not published by the providerNot enough data yetNot measured
Magnum V2 72B
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
101.2
From 6 benchmarks. A fair amount of data behind it.
Not enough data yet$2.006cache read $1.003$2.99216KNot enough data yetNot measured
Text Embedding 3 Small
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
101.2
From 117 benchmarks. Plenty of data behind it.
Not enough data yet$0.02β€”8KNot enough data yetNot measured
Text Embedding Ada 002
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
101.1
From 38 benchmarks. A fair amount of data behind it.
Not enough data yet$0.10β€”8KNot enough data yetNot measured
FLUX.2 [flex]
πŸ‡©πŸ‡ͺ
Maker headquartered in DE, according to the company's own website. The citation is on the model page.
101.0
From 151 benchmarks. A fair amount of data behind it.
Not enough data yetβ€”β€”Not published by the providerNot enough data yetNot measured
FLUX.2 [flex]
πŸ‡©πŸ‡ͺ
Maker headquartered in DE, according to the company's own website. The citation is on the model page.
flux-2-flex-image-to-image Β· image-to-image
101.0
From 151 benchmarks. A fair amount of data behind it.
Not enough data yetβ€”β€”Not published by the providerNot enough data yetNot measured
BGE M3
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.
100.9
From 136 benchmarks. Plenty of data behind it.
Not enough data yet$0.01β€”8KNot enough data yetNot measured
Ministral 3B
πŸ‡«πŸ‡·
Maker headquartered in FR, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
mistralai/ministral-3b-2512 Β· 2512
100.6
From 24 benchmarks. Plenty of data behind it.
Not enough data yet$0.10cache read $0.01$0.10131K
38.6
Poor value (38.6). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measured
Ministral 3B
πŸ‡«πŸ‡·
Maker headquartered in FR, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
mistralai/ministral-3b-2512 Β· 2512 Β· int4 Β· GLB
Inference by Kitani
100.6
From 24 benchmarks. Plenty of data behind it.
Not enough data yet$0.0945cache read $0.0473$0.0945Not published by the provider
38.6
Poor value (38.6). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measured
FLUX.2 [max]
πŸ‡©πŸ‡ͺ
Maker headquartered in DE, according to the company's own website. The citation is on the model page.
100.5
From 170 benchmarks. A fair amount of data behind it.
Not enough data yetβ€”β€”Not published by the providerNot enough data yetNot measured
FLUX.2 [max]
πŸ‡©πŸ‡ͺ
Maker headquartered in DE, according to the company's own website. The citation is on the model page.
flux-2-max-image-to-image Β· image-to-image
100.5
From 170 benchmarks. A fair amount of data behind it.
Not enough data yetβ€”β€”Not published by the providerNot enough data yetNot measured
Llama 3.1 Nemotron 70B Instruct HF
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.
100.2
From 6 benchmarks. A fair amount of data behind it.
Not enough data yet$0.357cache read $0.1785$0.40816KNot enough data yetNot measured
GPT Image 1
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
100.1
From 45 benchmarks. A fair amount of data behind it.
Not enough data yetβ€”β€”Not published by the providerNot enough data yetNot measured
FLUX.2 [pro]
πŸ‡©πŸ‡ͺ
Maker headquartered in DE, according to the company's own website. The citation is on the model page.
99.9
From 155 benchmarks. A fair amount of data behind it.
Not enough data yetβ€”β€”Not published by the providerNot enough data yetNot measured
FLUX.2 [pro]
πŸ‡©πŸ‡ͺ
Maker headquartered in DE, according to the company's own website. The citation is on the model page.
flux-2-pro-image-to-image Β· image-to-image
99.9
From 155 benchmarks. A fair amount of data behind it.
Not enough data yetβ€”β€”Not published by the providerNot enough data yetNot measured
Doubao Seedream 4.5
Newer version
Newer versionSeedream 5.5We compared the version names in our catalogue.
seedream-v4.5
98.2
From 177 benchmarks. A fair amount of data behind it.
Not enough data yetβ€”β€”Not published by the providerNot enough data yetNot measured
Doubao Seedream 4.5
Newer version
Newer versionSeedream 5.5We compared the version names in our catalogue.
seedream-v4.5-sequential Β· sequential
98.2
From 177 benchmarks. A fair amount of data behind it.
Not enough data yetβ€”β€”Not published by the providerNot enough data yetNot measured
Mistral Nemo
πŸ‡«πŸ‡·
Maker headquartered in FR, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
mistralai/mistral-nemo-instruct-2407
98.1
From 85 benchmarks. Plenty of data behind it.
Not enough data yet$0.1003cache read $0.0502$0.120716K
52.2
Fair value (52.2). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate.
113.9
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Mistral Nemo
πŸ‡«πŸ‡·
Maker headquartered in FR, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
mistralai/mistral-nemo-instruct-2407 Β· int4 Β· GLB
Inference by Kitani
98.1
From 85 benchmarks. Plenty of data behind it.
Not enough data yet$0.0948cache read $0.0474$0.1141Not published by the provider
52.2
Fair value (52.2). Costs per task about what its benchmark results predict. Little data behind it, so treat it as a rough estimate.
113.9
Epoch Capabilities IndexSource: Epoch AI Benchmarking Hub API
Mixtral 8x22B Instruct v0.1
πŸ‡«πŸ‡·
Maker headquartered in FR, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
96.8
From 154 benchmarks. Plenty of data behind it.
β‰ˆ 76Β Β±12
Coding IQ 75.7, 95% interval 64–88. 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.
$2.00cache read $0.20$6.0066K
33.1
Poor value (33.1). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measured
Llama 3 70b Instruct Abliterated v3.5
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
failspy/Meta-Llama-3-70B-Instruct-abliterated-v3.5
95.8
From 6 benchmarks. A fair amount of data behind it.
Not enough data yet$0.70cache read $0.35$0.708KNot enough data yetNot measured
Llama 3.1 8B Instruct
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
Newer version
Newer versionLlama-3.3-8B-InstructWe compared the version names in our catalogue.
95.6
From 343 benchmarks. Plenty of data behind it.
70Β Β±9
Coding IQ 69.9, 95% interval 61–79. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.02cache read $0.01$0.04131K
39.5
Poor value (39.5). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measured
Llama 3.1 8B Instruct
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
Newer version
Newer versionLlama-3.3-8B-InstructWe compared the version names in our catalogue.
Meta-Llama-3-1-8B-Instruct-FP8 Β· fp8
95.6
From 343 benchmarks. Plenty of data behind it.
70Β Β±9
Coding IQ 69.9, 95% interval 61–79. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.02cache read $0.01$0.03128K
39.5
Poor value (39.5). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measured
Llama 3.1 8B Instruct
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
Newer version
Newer versionLlama-3.3-8B-InstructWe compared the version names in our catalogue.
πŸ‡ΊπŸ‡Έ US
Inference by Awsbedrock
95.6
From 343 benchmarks. Plenty of data behind it.
70Β Β±9
Coding IQ 69.9, 95% interval 61–79. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.231$0.231Not published by the provider
39.5
Poor value (39.5). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measured
Llama 3.1 8B Instruct
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
Newer version
Newer versionLlama-3.3-8B-InstructWe compared the version names in our catalogue.
bf16 Β· πŸ‡ΊπŸ‡Έ US
Inference by Coreweave
95.6
From 343 benchmarks. Plenty of data behind it.
70Β Β±9
Coding IQ 69.9, 95% interval 61–79. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.231$0.231Not published by the provider
39.5
Poor value (39.5). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measured
Llama 3.1 8B Instruct
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
Newer version
Newer versionLlama-3.3-8B-InstructWe compared the version names in our catalogue.
fp8 Β· πŸ‡ΊπŸ‡Έ US
Inference by DeepInfra
95.6
From 343 benchmarks. Plenty of data behind it.
70Β Β±9
Coding IQ 69.9, 95% interval 61–79. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.021$0.042Not published by the provider
39.5
Poor value (39.5). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measured
Llama 3.1 8B Instruct
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
Newer version
Newer versionLlama-3.3-8B-InstructWe compared the version names in our catalogue.
πŸ‡ΊπŸ‡Έ US
Inference by Groq
95.6
From 343 benchmarks. Plenty of data behind it.
70Β Β±9
Coding IQ 69.9, 95% interval 61–79. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.0525cache read $0.0263$0.084Not published by the provider
39.5
Poor value (39.5). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measured
Llama 3.1 8B Instruct
πŸ‡ΊπŸ‡Έ
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
Newer version
Newer versionLlama-3.3-8B-InstructWe compared the version names in our catalogue.
int4 Β· GLB
Inference by Kitani
95.6
From 343 benchmarks. Plenty of data behind it.
70Β Β±9
Coding IQ 69.9, 95% interval 61–79. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.
$0.0514cache read $0.0257$0.0803Not published by the provider
39.5
Poor value (39.5). Costs more per task than its benchmark results predict. A fair amount of data behind it.
Not measured
Phi 4 Mini
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.
phi-4-mini-instruct
95.6
From 41 benchmarks. Plenty of data behind it.
Not enough data yet$0.17cache read $0.085$0.688K
40.9
Poor value (40.9). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measured
Phi 4 Mini
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.
phi-4-mini-instruct Β· int4 Β· GLB
Inference by Kitani
95.6
From 41 benchmarks. Plenty of data behind it.
Not enough data yet$0.1607cache read $0.0803$0.693Not published by the provider
40.9
Poor value (40.9). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Not measured
MN 12b Mag Mell R1
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
95.2
From 6 benchmarks. A fair amount of data behind it.
Not enough data yet$0.493cache read $0.2465$0.49316KNot enough data yetNot measured
Neural Daredevil 8B abliterated
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
95.0
From 6 benchmarks. A fair amount of data behind it.
Not enough data yet$0.44cache read $0.22$0.448KNot enough data yetNot measured
Neural Daredevil 8B abliterated
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
int4 Β· GLB
Inference by Kitani
95.0
From 6 benchmarks. A fair amount of data behind it.
Not enough data yet$0.441cache read $0.2205$0.441Not published by the providerNot enough data yetNot measured
Doubao Seedream 4-0
Newer version
Newer versionSeedream 5.5We compared release dates and version names in our catalogue.
seedream-v4
93.4
From 182 benchmarks. A fair amount of data behind it.
Not enough data yetβ€”β€”Not published by the providerNot enough data yetNot measured
FLUX.2 [dev]
πŸ‡©πŸ‡ͺ
Maker headquartered in DE, according to the company's own website. The citation is on the model page.
92.4
From 183 benchmarks. A fair amount of data behind it.
Not enough data yetβ€”β€”Not published by the providerNot enough data yetNot measured
FLUX.2 [dev]
πŸ‡©πŸ‡ͺ
Maker headquartered in DE, according to the company's own website. The citation is on the model page.
flux-2-dev-image-to-image Β· image-to-image
92.4
From 183 benchmarks. A fair amount of data behind it.
Not enough data yetβ€”β€”Not published by the providerNot enough data yetNot measured
FLUX.2 [dev]
πŸ‡©πŸ‡ͺ
Maker headquartered in DE, according to the company's own website. The citation is on the model page.
flux-2-dev-lora-image-to-image Β· image-to-image+lora
92.4
From 183 benchmarks. A fair amount of data behind it.
Not enough data yetβ€”β€”Not published by the providerNot enough data yetNot measured
FLUX.2 [dev]
πŸ‡©πŸ‡ͺ
Maker headquartered in DE, according to the company's own website. The citation is on the model page.
flux-2-dev-lora Β· lora
92.4
From 183 benchmarks. A fair amount of data behind it.
Not enough data yetβ€”β€”Not published by the providerNot enough data yetNot measured
β‰ˆ 92
From 13 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yetβ€”β€”Not published by the providerNot enough data yetNot measured

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

How this data works