Free models and APIs on NVIDIA Build in 2026
NVIDIA Build provides free serverless inference for development. This catalog is refreshed directly from the authenticated NIM Models API.
63 modelsup to 1M contextbest: Step 5 PreviewUpdated:
NVIDIA Build
Online at the last check
Free serverless inference for development and prototyping with a developer account; not intended for production traffic.
- Base URL
https://integrate.api.nvidia.com/v1- Compatibility
- OpenAI Chat Completions
- Card
- Not required
- Limits
- Free serverless inference for development; limits and model availability may change.
| # | Model ↕ | AA Index ↕ | Context ↕ | Size ↕ | Limit | |
|---|---|---|---|---|---|---|
| 1 | 
Step 5 Preview
stepfun-ai/step-5-preview |
43.7 | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 2 | 
GLM 5.3 Flash
z-ai/glm-5.3-flash |
41.8 | — | Mid-size | Free serverless inference for development; limits and model availability may change. | Open |
| 3 | 
DeepSeek V4.1 Flash
deepseek-ai/deepseek-v4.1-flash |
39.5 | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 4 | 
Kimi K2.6
moonshotai/kimi-k2.6 |
27.0 | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 5 | 
Nemotron 3 Ultra
nvidia/nemotron-3-ultra-550b-a55b |
22.9 | 1M | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 6 | 
Gemma 4 31B
google/gemma-4-31b-it |
19.0 | — | Mid-size | Free serverless inference for development; limits and model availability may change. | Open |
| 7 |
Nemotron 3.5 Lightning
nvidia/nemotron-3.5-lightning-30b-a3b |
12.9 | — | Mid-size | Free serverless inference for development; limits and model availability may change. | Open |
| 8 | 
MiniMax M3.1 Flash
minimaxai/minimax-m3.1-flash |
— | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 9 |
Nemotron 3 Nano Omni
nvidia/nemotron-3-nano-omni-30b-a3b-reasoning |
10.3 | — | Mid-size | Free serverless inference for development; limits and model availability may change. | Open |
| 10 |
Nemotron 3 Super
nvidia/nemotron-3-super-120b-a12b |
12.8 | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 11 | 
gpt-oss-20b
openai/gpt-oss-20b |
10.0 | — | Mid-size | Free serverless inference for development; limits and model availability may change. | Open |
| 12 |
Nemotron 3.5 Super VL 120B
nvidia/nemotron-3.5-super-vl-120b-a12b |
— | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 13 |
Palmyra Creative 122B
writer/palmyra-creative-122b |
— | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 14 | 
Mistral Nemo
nv-mistralai/mistral-nemo-12b-instruct |
— | — | Mid-size | Free serverless inference for development; limits and model availability may change. | Open |
| 15 |
Nemotron 3 Nano
nvidia/nemotron-nano-3-30b-a3b |
8.9 | — | Mid-size | Free serverless inference for development; limits and model availability may change. | Open |
| 16 |
Laguna XS 2.1
poolside/laguna-xs-2.1 |
— | — | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 17 |
Muse Glimmer 30B
meta/muse-glimmer-30b |
— | — | Mid-size | Free serverless inference for development; limits and model availability may change. | Open |
| 18 | 
Llama 3.2 11B Vision
meta/llama-3.2-11b-vision-instruct |
5.4 | — | Mid-size | Free serverless inference for development; limits and model availability may change. | Open |
| 19 |
Nemotron 3.5 Bolt E4b
nvidia/nemotron-3.5-bolt-e4b |
— | — | Mid-size | Free serverless inference for development; limits and model availability may change. | Open |
| 20 |
Llama 3.2 90B Vision
meta/llama-3.2-90b-vision-instruct |
6.4 | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 21 |
Llama 3.1 Nemotron Ultra 253B
nvidia/llama-3.1-nemotron-ultra-253b-v1 |
7.5 | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 22 |
Mixtral 8x22b
mistralai/mixtral-8x22b-v0.1 |
— | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 23 |
Gemma 3 4B
google/gemma-3-4b-it |
4.8 | — | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 24 |
CodeLlama 70B
meta/codellama-70b |
— | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 25 |
Llama 2 70B
meta/llama2-70b |
— | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 26 |
Llama 3 ChatQA 1.5 70B
nvidia/llama3-chatqa-1.5-70b |
— | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 27 |
Mistral Large 4 1t A52B
mistralai/mistral-large-4-1t-a52b |
— | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 28 |
Nemotron 4 340B
nvidia/nemotron-4-340b-instruct |
— | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 29 |
Palmyra Fin 70B 32K
writer/palmyra-fin-70b-32k |
— | 33K | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 30 |
Palmyra Med 70B
writer/palmyra-med-70b |
— | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 31 |
Palmyra Med 70B 32K
writer/palmyra-med-70b-32k |
— | 33K | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 32 |
Yi Large
01-ai/yi-large |
— | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 33 |
Cosmos Reason2 8B
nvidia/cosmos-reason2-8b |
— | — | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 34 |
Mistral Nemo Minitron 8B 8K
nvidia/mistral-nemo-minitron-8b-8k-instruct |
— | 8K | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 35 |
Gemma 3 12B
google/gemma-3-12b-it |
3.8 | — | Mid-size | Free serverless inference for development; limits and model availability may change. | Open |
| 36 |
Mistral 7B
mistralai/mistral-7b-instruct-v0.3 |
5.0 | — | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 37 |
Codestral 22B
mistralai/codestral-22b-instruct-v0.1 |
— | — | Mid-size | Free serverless inference for development; limits and model availability may change. | Open |
| 38 |
Granite 34B Code
ibm/granite-34b-code-instruct |
— | — | Mid-size | Free serverless inference for development; limits and model availability may change. | Open |
| 39 |
Llama 3.1 Nemotron 51B
nvidia/llama-3.1-nemotron-51b-instruct |
— | — | Mid-size | Free serverless inference for development; limits and model availability may change. | Open |
| 40 |
Neva 22B
nvidia/neva-22b |
— | — | Mid-size | Free serverless inference for development; limits and model availability may change. | Open |
| 41 | 
Phi 3.5 MoE
microsoft/phi-3.5-moe-instruct |
— | — | Mid-size | Free serverless inference for development; limits and model availability may change. | Open |
| 42 |
StarCoder2 15B
bigcode/starcoder2-15b |
— | — | Mid-size | Free serverless inference for development; limits and model availability may change. | Open |
| 43 |
Llama 3.1 Nemotron 70B
nvidia/llama-3.1-nemotron-70b-instruct |
6.9 | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 44 |
Gemma 2B
google/gemma-2b |
— | — | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 45 |
Mistral Large 2
mistralai/mistral-large-2-instruct |
7.6 | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 46 | 
Jamba 1.5 Large
ai21labs/jamba-1.5-large-instruct |
6.0 | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 47 |
Mistral Large
mistralai/mistral-large |
5.8 | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 48 |
DBRX Instruct
databricks/dbrx-instruct |
5.3 | — | Frontier | Free serverless inference for development; limits and model availability may change. | Open |
| 49 |
CodeGemma 1.1 7B
google/codegemma-1.1-7b |
— | — | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 50 |
CodeGemma 7B
google/codegemma-7b |
— | — | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 51 |
DeepSeek Coder 6 7B
deepseek-ai/deepseek-coder-6.7b-instruct |
— | — | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 52 |
Fuyu 8B
adept/fuyu-8b |
— | — | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 53 |
Granite 3.0 3B A800M
ibm/granite-3.0-3b-a800m-instruct |
— | — | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 54 |
Granite 3.0 8B
ibm/granite-3.0-8b-instruct |
— | — | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 55 |
Granite 8B Code
ibm/granite-8b-code-instruct |
— | — | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 56 |
Kosmos 2
microsoft/kosmos-2 |
— | — | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 57 |
Phi 3 Vision 128K
microsoft/phi-3-vision-128k-instruct |
— | 131K | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 58 |
RecurrentGemma 2B
google/recurrentgemma-2b |
— | — | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 59 |
SEA-LION 7B
aisingapore/sea-lion-7b-instruct |
— | — | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 60 |
Vila
nvidia/vila |
— | — | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 61 |
Zamba2 7B Instruct
zyphra/zamba2-7b-instruct |
— | — | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 62 |
Riva Translate 4B Instruct V2
nvidia/riva-translate-4b-instruct-v2 |
— | — | Small | Free serverless inference for development; limits and model availability may change. | Open |
| 63 |
Riva Translate 4B Instruct
nvidia/riva-translate-4b-instruct |
— | — | Small | Free serverless inference for development; limits and model availability may change. | Open |
No models match your filters.
Benchmarks: Artificial Analysis (September 24, 2026). A dash means the model is not on their leaderboard; we never estimate scores ourselves.
FAQ
Is NVIDIA Build free?
Free serverless inference for development and prototyping with a developer account; not intended for production traffic. 63 models are currently free.
Is NVIDIA Build compatible with the OpenAI API?
Yes — use the base URL https://integrate.api.nvidia.com/v1 with any OpenAI-compatible client or SDK.
Does NVIDIA Build require a card?
No, free models work without a card.