17 models tracked · known for Qwen3.6-35B-A3B-NVFP4, Gemma-4-31B-IT-NVFP4, DeepSeek-V4-Flash
Claims drawn from cited facts, not live model generation.
NVIDIA is a lab that publishes news and tutorials for developers, data scientists, and IT admins.
3 cited facts
NVIDIA has shipped 17 models as tracked in this graph. Its 100 recent blog announcements signal an actively communicating lab. Together, this mix reflects a research-driven lab that is also productively shipping, rather than quietly building in the background.
5 cited facts
26 benchmark scores across this lab’s models, tagged by evidence status.
| Model | Best score | Claim status |
|---|---|---|
| Kimi K2.6 | 96.11 | independently reproduced |
| Nemotron-4 15B | 76.53 | self-reported |
| Nemotron 3 Ultra | 43.45 | unverified |
| Qwen3.6-35B-A3B-NVFP4 | — | unverified |
| Gemma-4-31B-IT-NVFP4 | — | unverified |
| NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4 | — | unverified |
| Gemma-4-26B-A4B-NVFP4 | — | unverified |
| Qwen3.6-27B-NVFP4 | — | unverified |
| GLM-5.2-NVFP4 | — | unverified |
| Kimi-K2.7-Code-NVFP4 | — | unverified |
| NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 | — | unverified |
| diffusiongemma-26B-A4B-it-NVFP4 | — | unverified |
| NVIDIA-Nemotron-3-Super-120B-A12B-BF16 | — | unverified |
| NVIDIA-Nemotron-3-Nano-30B-A3B-FP8 | — | unverified |
| NVIDIA-Nemotron-3-Nano-30B-A3B-NVFP4 | — | unverified |
| NVIDIA-Nemotron-3-Nano-4B-BF16 | — | unverified |
Source facts, citations, and refresh stamp for this record.
Sources: epoch_benchmarks, huggingface_models, lab_blog_rss