Skip to content

Listings

All listings: NanoGPT

Every listing for the models on page 12 of 40 of NanoGPT, including each model's other versions and routes. Back to NanoGPT

Sold by NanoGPT

38 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 TokensValue
Seedream 3seedream-v3
≈ 92
From 166 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet——Not published by the providerNot enough data yet
FLUX.2 [klein] 9B
🇩🇪
Maker headquartered in DE, according to the company's own website. The citation is on the model page.
90.8
From 160 benchmarks. A fair amount of data behind it.
Not enough data yet——Not published by the providerNot enough data yet
FLUX.2 [klein] 4B
🇩🇪
Maker headquartered in DE, according to the company's own website. The citation is on the model page.
89.9
From 161 benchmarks. A fair amount of data behind it.
Not enough data yet——Not published by the providerNot enough data yet
Qwen Image
🇨🇳
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
89.7
From 177 benchmarks. A fair amount of data behind it.
Not enough data yet——Not published by the providerNot enough data yet
Qwen Image
🇨🇳
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
qwen-image-2512 · 2512
89.7
From 177 benchmarks. A fair amount of data behind it.
Not enough data yet——Not published by the providerNot enough data yet
Reka Edgereka/reka-edge-2603 · 2603
89.3
From 51 benchmarks. A fair amount of data behind it.
Not enough data yet$0.105$0.10516K
38.1
Poor value (38.1). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Reka Edgereka/reka-edge-2603 · 2603 · UNK
Inference by Reka AI
89.3
From 51 benchmarks. A fair amount of data behind it.
Not enough data yet$0.105$0.105Not published by the provider
38.1
Poor value (38.1). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Z Image Turbo
🇨🇳
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.
≈ 89
From 147 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet——Not published by the providerNot enough data yet
Z Image Turbo
🇨🇳
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-image-turbo-image-to-image · image-to-image
≈ 89
From 147 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet——Not published by the providerNot enough data yet
Z Image Turbo
🇨🇳
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-image-turbo-lora · lora
≈ 89
From 147 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet——Not published by the providerNot enough data yet
Hunyuan Image V3
🇨🇳
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
hunyuan-image-3
89.2
From 42 benchmarks. A fair amount of data behind it.
Not enough data yet——Not published by the providerNot enough data yet
Veo 3
🇺🇸
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Newer version
Newer versionVeo 3.1We compared release dates and version names in our catalogue.
veo3-video · audio+duration-8s
≈ 89
From 41 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet——Not published by the providerNot enough data yet
Veo 3
🇺🇸
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Newer version
Newer versionVeo 3.1We compared release dates and version names in our catalogue.
veo3-video · duration-8s+no-audio
≈ 89
From 41 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet——Not published by the providerNot enough data yet
BGE Large En V1.5
🇨🇳
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.
87.1
From 155 benchmarks. Plenty of data behind it.
Not enough data yet$0.01—512Not enough data yet
LTX-2 19B1080p
≈ 87
From 5 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet——Not published by the providerNot enough data yet
LTX-2 19B480p
≈ 87
From 5 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet——Not published by the providerNot enough data yet
LTX-2 19B720p
≈ 87
From 5 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet——Not published by the providerNot enough data yet
BGE Base En V1.5
🇨🇳
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.
86.7
From 152 benchmarks. Plenty of data behind it.
Not enough data yet$0.005—512Not enough data yet
Llama 3.2 3B
🇺🇸
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.
meta-llama/llama-3.2-3b-instruct
86.5
From 255 benchmarks. Plenty of data behind it.
≈ 76 ±14
Coding IQ 75.7, 95% interval 62–89. 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.
$0.05cache read $0.025$0.33131K
44.3
Poor value (44.3). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Llama 3.2 3B
🇺🇸
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.
meta-llama/llama-3.2-3b-instruct · int4 · GLB
Inference by Kitani
86.5
From 255 benchmarks. Plenty of data behind it.
≈ 76 ±14
Coding IQ 75.7, 95% interval 62–89. 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.
$0.0289cache read $0.0145$0.0466Not published by the provider
44.3
Poor value (44.3). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
Llama 3.2 3B
🇺🇸
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.
meta-llama/llama-3.2-3b-instruct · bf16 · 🇺🇸 US
Inference by Parasail
86.5
From 255 benchmarks. Plenty of data behind it.
≈ 76 ±14
Coding IQ 75.7, 95% interval 62–89. 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.
$0.0525$0.3465Not published by the provider
44.3
Poor value (44.3). Costs more per task than its benchmark results predict. Little data behind it, so treat it as a rough estimate.
GPT-4o mini TTS
🇺🇸
Maker headquartered in US, according to the company's own website. The citation is on the model page.
≈ 85
From 16 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet——Not published by the providerNot enough data yet
GPT-4o mini TTS
🇺🇸
Maker headquartered in US, according to the company's own website. The citation is on the model page.
gpt-4o-mini-tts-2025-03-20 · 2025-03-20
≈ 85
From 16 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet——Not published by the providerNot enough data yet
GPT-4o mini TTS
🇺🇸
Maker headquartered in US, according to the company's own website. The citation is on the model page.
gpt-4o-mini-tts-2025-12-15 · 2025-12-15
≈ 85
From 16 benchmarks. Little data behind it, so treat it as a rough estimate.
Not enough data yet——Not published by the providerNot enough data yet
GLM-Image
🇨🇳
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.
83.6
From 152 benchmarks. A fair amount of data behind it.
Not enough data yet——Not published by the providerNot enough data yet
Flux Kontext Dev
🇩🇪
Maker headquartered in DE, according to the company's own website. The citation is on the model page.
Open weights. The weights are publicly available, according to the provider's model list or a public Hugging Face repo.
runware:106@1
80.4
From 23 benchmarks. A fair amount of data behind it.
Not enough data yet——Not published by the providerNot enough data yet
80.1
From 186 benchmarks. A fair amount of data behind it.
Not enough data yet——Not published by the providerNot enough data yet
Ling 3.0 Tiny
🇨🇳
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$0.015cache read $0.003$0.045262KNot enough data yet
Ling 3.0 Tiny
🇨🇳
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.
inclusionai/ling-3.0-tiny:thinking · thinking
Not enough data yetNot enough data yet$0.015cache read $0.003$0.045262KNot enough data yet
Nemotron 3.5 Embed 1b
🇺🇸
Maker headquartered in US, according to the company's own website. The citation is on the model page.
nvidia/Nemotron-3.5-Embed-1B-BF16 · bf16
Not enough data yetNot enough data yet$0.015—33KNot enough data yet
Not enough data yetNot enough data yet$0.02cache read $0.01$0.1616KNot enough data yet
Not enough data yetNot enough data yet$0.02cache read $0.01$0.168KNot enough data yet
Nex N2.5 Mini
🇨🇳
Maker headquartered in CN, according to the company's own website. The citation is on the model page.
Not enough data yetNot enough data yet$0.025cache read $0.0025$0.10262KNot enough data yet
Schematron V2 Turbo
🇺🇸
Maker headquartered in US, according to the company's own website. The citation is on the model page.
Not enough data yetNot enough data yet$0.03cache read $0.015$0.15128KNot enough data yet
Synth 2.5 Flash Previewsynth-2.5-flash
Not enough data yetNot enough data yet$0.03cache read $0.015$0.1030KNot enough data yet
Synth 2.5 Flash Previewsynth-2.5-flash · UNK
Inference by Lucidity
Not enough data yetNot enough data yet$0.0315$0.105Not published by the providerNot enough data yet
Nemotron 3.5 Embed 8b
🇺🇸
Maker headquartered in US, according to the company's own website. The citation is on the model page.
nvidia/Nemotron-3.5-Embed-8B-BF16 · bf16
Not enough data yetNot enough data yet$0.035—33KNot enough data yet
Not enough data yetNot enough data yet$0.04cache read $0.004$0.15260KNot enough data yet

How this data works