NVIDIA NIM gallery information¶
This table combines all NVIDIA NIM model information, including model names, types, chat model IDs, playground support, platform support, and documentation links.
| NIM | Type | Chat model ID | Supported in playground | Platform support |
|---|---|---|---|---|
| codellama-13b-instruct | Text Generation | codellama/codellama-13b-instruct | Yes | Cloud, 11.1, 11.2 |
| codellama-34b-instruct | Text Generation | codellama/codellama-34b-instruct | Yes | Cloud, 11.1, 11.2 |
| codellama-70b-instruct | Text Generation | codellama/codellama-70b-instruct | Yes | Cloud, 11.1, 11.2 |
| deepseek-r1-distill-llama-8b | Text Generation | deepseek-ai/deepseek-r1-distill-llama-8b | Yes | Cloud, 11.1, 11.2 |
| deepseek-r1-distill-qwen-7b | Text Generation | deepseek-ai/deepseek-r1-distill-qwen-7b | Yes | Cloud, 11.1, 11.2 |
| deepseek-r1-distill-qwen-14b | Text Generation | deepseek-ai/deepseek-r1-distill-qwen-14b | Yes | Cloud, 11.1, 11.2 |
| deepseek-r1-distill-qwen-32b | Text Generation | deepseek-ai/deepseek-r1-distill-qwen-32b | Yes | Cloud, 11.1, 11.2 |
| gemma-2-2b-instruct | Text Generation | google/gemma-2-2b-instruct | Yes | Cloud, 11.1, 11.2 |
| gemma-2-9b-it | Text Generation | google/gemma-2-9b-it | Yes | Cloud, 11.1, 11.2 |
| gpt-oss-120b | Text Generation | openai/gpt-oss-120b | Yes | Cloud, 11.2 |
| gpt-oss-20b | Text Generation | openai/gpt-oss-20b | Yes | Cloud, 11.2 |
| llama-2-13b-chat | Text Generation | meta/llama-2-13b-chat | Yes | Cloud, 11.1, 11.2 |
| llama-2-7b-chat | Text Generation | meta/llama-2-7b-chat | Yes | Cloud, 11.1, 11.2 |
| llama-2-70b-chat | Text Generation | meta/llama-2-70b-chat | No | 11.1, 11.2 |
| llama-3-sqlcoder-8b | Text Generation | defog/llama-3-sqlcoder-8b | Yes | Cloud, 11.1, 11.2 |
| llama-3-swallow-70b-instruct-v0.1 | Text Generation | tokyotech-llm/llama-3-swallow-70b-instruct-v0.1 | No | 11.1, 11.2 |
| llama-3-taiwan-70b-instruct | Text Generation | yentinglin/llama-3-taiwan-70b-instruct | No | 11.1, 11.2 |
| llama-3.1-70b-instruct | Text Generation | meta/llama-3.1-70b-instruct | Yes | Cloud, 11.1, 11.2 |
| llama-3.1-8b-instruct | Text Generation | meta/llama-3.1-8b-instruct | Yes | Cloud, 11.1, 11.2 |
| llama-3.1-8b-instruct-pb24h2 | Text Generation | meta/llama-3.1-8b-instruct-pb24h2 | Yes | Cloud, 11.1, 11.2 |
| llama-3.1-70b-instruct-pb24h2 | Text Generation | meta/llama-3.1-70b-instruct-pb24h2 | Yes | Cloud, 11.1, 11.2 |
| llama-3.1-nemotron-nano-8b-v1 | Text Generation | nvidia/llama-3.1-nemotron-nano-8b-v1 | Yes | Cloud, 11.1, 11.2 |
| llama-3.1-nemotron-70b-instruct | Text Generation | nvidia/llama-3.1-nemotron-70b-instruct | Yes | Cloud, 11.1, 11.2 |
| llama-3.1-nemotron-ultra-253b-v1 | Text Generation | nvidia/llama-3.1-nemotron-ultra-253b-v1 | No | 11.1, 11.2 |
| llama-3.1-swallow-70b-instruct-v0.1 | Text Generation | tokyotech-llm/llama-3.1-swallow-70b-instruct-v0.1 | Yes | Cloud, 11.1, 11.2 |
| llama-3.2-1b-instruct | Text Generation | meta/llama-3.2-1b-instruct | Yes | Cloud, 11.1, 11.2 |
| llama-3.2-3b-instruct | Text Generation | meta/llama-3.2-3b-instruct | Yes | Cloud, 11.1, 11.2 |
| llama-3.2-11b-vision-instruct | Text Generation | meta/llama-3.2-11b-vision-instruct | Yes | Cloud, 11.1, 11.2 |
| llama-3.2-90b-vision-instruct | Text Generation | meta/llama-3.2-90b-vision-instruct | No | 11.1, 11.2 |
| llama-3.3-70b-instruct | Text Generation | meta/llama-3.3-70b-instruct | Yes | Cloud, 11.1, 11.2 |
| llama-3.3-nemotron-super-49b-v1 | Text Generation | nvidia/llama-3.3-nemotron-super-49b-v1 | Yes | Cloud, 11.1, 11.2 |
| llama-3.3-nemotron-super-49b-v1.5 | Text Generation | nvidia/llama-3-3-nemotron-super-49b-v1-5 | Yes | Cloud, 11.2 |
| llama-4-scout-17b-16e-instruct | Text Generation | meta/llama-4-scout-17b-16e-instruct | Yes | 11.2 |
| llama3-70b-instruct | Text Generation | meta/llama3-70b-instruct | No | 11.1, 11.2 |
| llama3-8b-instruct | Text Generation | meta/llama3-8b-instruct | Yes | Cloud, 11.1, 11.2 |
| mistral-7b-instruct-v0.3 | Text Generation | mistralai/mistral-7b-instruct-v0.3 | Yes | Cloud, 11.1, 11.2 |
| mistral-nemo-12b-instruct | Text Generation | mistral-nemo-12b-instruct | Yes | Cloud, 11.1, 11.2 |
| mistral-nemo-minitron-8b-8k-instruct | Text Generation | nv-mistralai/mistral-nemo-minitron-8b-8k-instruct | Yes | Cloud, 11.1, 11.2 |
| mixtral-8x7b-instruct-v01 | Text Generation | mistralai/mixtral-8x7b-instruct-v0.1 | Yes | Cloud, 11.1, 11.2 |
| mixtral-8x22b-instruct-v01 | Text Generation | mistralai/mixtral-8x22b-instruct-v01 | No | 11.1, 11.2 |
| nvidia-nemotron-nano-9b-v2 | Text Generation | nvidia/nvidia-nemotron-nano-9b-v2 | Yes | 11.1, 11.2 |
| phi-3-mini-4k-instruct | Text Generation | microsoft/phi-3-mini-4k-instruct | Yes | Cloud, 11.1, 11.2 |
| qwen-2.5-7b-instruct | Text Generation | qwen/qwen-2.5-7b-instruct | Yes | Cloud, 11.1, 11.2 |
| qwen3-32b | Text Generation | qwen/qwen3-32b | Yes | 11.2 |
| qwen3-next-80b-a3b-thinking | Text Generation | qwen/qwen3-next-80b-a3b-thinking | Yes | 11.2 |
| starcoder2-7b | Text Generation | bigcode/starcoder2-7b | Yes | Cloud, 11.1, 11.2 |
| cosmos-predict1-7b-text2world | Unstructured | - | - | 11.2 |
| cosmos-predict1-7b-video2world | Unstructured | - | - | 11.2 |
| cuopt | Unstructured | - | - | Cloud, 11.1, 11.2 |
| genmol | Unstructured | - | - | Cloud, 11.1, 11.2 |
| arctic-embed-l | Embedding/Unstructured | - | - | Cloud, 11.1, 11.2 |
| llama-3.1-nemotron-nano-vl-8b-v1 | Unstructured | - | - | 11.2 |
| llama-3.2-nv-embedqa-1b-v2 | Embedding/Unstructured | - | - | Cloud, 11.1, 11.2 |
| nv-embedqa-e5-v5 | Embedding/Unstructured | - | - | Cloud, 11.1, 11.2 |
| nv-embedqa-e5-v5-pb24h2 | Embedding/Unstructured | - | - | Cloud, 11.1, 11.2 |
| nv-embedqa-mistral-7b-v2 | Embedding/Unstructured | - | - | Cloud, 11.1, 11.2 |
| nvclip | Embedding/Unstructured | - | - | Cloud, 11.1, 11.2 |
| llama-3.2-nv-rerankqa-1b-v2 | Unstructured | - | - | Cloud, 11.1, 11.2 |
| molmim | Unstructured | - | - | Cloud, 11.1, 11.2 |
| nemoretriever-graphic-elements-v1 | Unstructured | - | - | Cloud, 11.1, 11.2 |
| nemoretriever-page-elements-v2 | Unstructured | - | - | Cloud, 11.1, 11.2 |
| nemoretriever-parse | Unstructured | - | - | Cloud, 11.1, 11.2 |
| nemoretriever-table-structure-v1 | Unstructured | - | - | Cloud, 11.1, 11.2 |
| nv-rerankqa-mistral-4b-v3 | Unstructured | - | - | Cloud, 11.1, 11.2 |
| openfold2 | Unstructured | - | - | 11.1, 11.2 |
| paddleocr | Unstructured | - | - | Cloud, 11.1, 11.2 |
| proteinmpnn | Unstructured | - | - | Cloud, 11.1, 11.2 |
| rfdiffusion | Unstructured | - | - | Cloud, 11.1, 11.2 |
| llama-3.1-nemoguard-8b-content-safety | Evaluation | - | - | Cloud, 11.1, 11.2 |
| llama-3.1-nemoguard-8b-topic-control | Evaluation | - | - | Cloud, 11.1, 11.2 |
| nemoguard-jailbreak-detect | Evaluation | - | - | Cloud, 11.1, 11.2 |
Feature considerations¶
- Chat model ID: For NIM model deployments, the chat model ID can be set to
datarobot-deployed-llmfor dynamic population, or hard-coded using the values in the table. - Playground support: Models marked as "No" in the playground support column are not supported in the playground.
-
Embedding/unstructured models with chat support:
The following embedding/unstructured models support both direct access endpoint and chat completions endpoint:- arctic-embed-l
- llama-3.2-nv-embedqa-1b-v2
- nv-embedqa-e5-v5
- nv-embedqa-e5-v5-pb24h2
- nv-embedqa-mistral-7b-v2
- nvclip
-
Evaluation metrics:
- llama-3.1-nemoguard-8b-topic-control: Stay on topic for input/output
- llama-3.1-nemoguard-8b-content-safety: Content safety
- nemoguard-jailbreak-detect: Jailbreak detection