Nvidia
Inference hostNvidia runs open and licensed models on its own hardware and sells them by usage, often faster or cheaper than the maker.
Models78
- Cheapest for
- 0 models
- Serves
- 4 model types
Monthly visits—
Models by type
Models on Nvidia
| Model | Input | Output | vs official | Context |
|---|---|---|---|---|
 Llama-3.2-90B-Vision-InstructMeta | Free/1M | Free/1M | -100% | — |
| Llama2 70B ChatMeta | Free/1M | Free/1M | -100% | — |
| Llama3 Chatqa 1.5 70BNVIDIAOnly here | Free/1M | Free/1M | Official | — |
 Mistral LargeMistral | Free/1M | Free/1M | -100% | — |
| Mistral Large 2 InstructMistral | Free/1M | Free/1M | -100% | — |
| Mistral Nemo 12B InstructMistralOnly here | Free/1M | Free/1M | — | — |
| Mistral Nemo Minitron 8B 8K InstructNVIDIAOnly here | Free/1M | Free/1M | Official | — |
| Mistral-7B-Instruct-v0.3MistralOnly here | Free/1M | Free/1M | — | — |
| Muse Glimmer 30BMeta | Free/1M | Free/1M | -100% | — |
| Nemotron 3.5 Embed 8BNVIDIAOnly here | Free/1M | Free/1M | Official | — |
| Nemotron 3.5 Super VL 120B A12BNVIDIA | Free/1M | Free/1M | Official | — |
| Nemotron 4 340B RewardNVIDIAOnly here | Free/1M | Free/1M | Official | — |
| Nemotron Nano 3 30B A3BNVIDIAOnly here | Free/1M | Free/1M | Official | — |
| Nemotron ParseNVIDIAOnly here | Free/1M | Free/1M | Official | — |
| Neva 22BNVIDIAOnly here | Free/1M | Free/1M | Official | — |
| Nv Embedqa Mistral 7B V2NVIDIAOnly here | Free/1M | Free/1M | Official | — |
| NvclipNVIDIAOnly here | Free/1M | Free/1M | Official | — |
 Palmyra Creative 122BWriterOnly here | Free/1M | Free/1M | — | — |
| Palmyra Fin 70B 32KWriterOnly here | Free/1M | Free/1M | — | — |
| Palmyra Med 70BWriterOnly here | Free/1M | Free/1M | — | — |
| Palmyra Med 70B 32KWriterOnly here | Free/1M | Free/1M | — | — |
 Phi 3 Vision 128K InstructMicrosoftOnly here | Free/1M | Free/1M | — | — |
 Recurrentgemma 2BGoogleOnly here | Free/1M | Free/1M | — | — |
| Riva Translate 4B InstructNVIDIAOnly here | Free/1M | Free/1M | Official | — |
 Sea Lion 7B InstructAI SingaporeOnly here | Free/1M | Free/1M | — | — |
 Step 5 PreviewStepFun | Free/1M | Free/1M | -100% | — |
| VilaNVIDIAOnly here | Free/1M | Free/1M | Official | — |
 Zamba2 7B InstructzyphraOnly here | Free/1M | Free/1M | — | — |
Nvidia at a glance
78 models in total: 63 text, 9 embedding, 5 moderation, 1 video. GLM-5.3 has the largest context window, 1.05M tokens. The most searched is GLM-5.3, with about 276K Google searches a month. The newest is MiniMax-M3.1-Flash, released Oct 17, 2026.
Largest context
GLM-5.31.05M
DeepSeek V4.1 Flash1.05M- GLM-5.3-Flash1.05M
MiniMax-M3.1-Flash1.05M- Step 5 Preview1.05M
Most searched
- GLM-5.3276K
- DeepSeek V4.1 Flash138K
- GLM-5.3-Flash121K
- MiniMax-M3.1-Flash29K
- GPT OSS 20B12K
Newest
- MiniMax-M3.1-FlashOct 17, 2026
- Nemotron 3.5 Super VL 120B A12BOct 15, 2026
- Nemotron 3.5 Embed 1BOct 7, 2026
- Nemotron 3.5 Embed 8BOct 7, 2026
- Step 5 PreviewSep 20, 2026