NVIDIA
Llama-3.3-Nemotron-Super-49B-v1.5 is a 49B-parameter, English-centric reasoning/chat model derived from Meta’s Llama-3.3-70B-Instruct with a 128K context. It’s post-trained for agentic workflows (RAG, tool calling) via SFT across math, code, science, and...
Running this yourself: likely needs a rented cloud gpu.
---
Quality Score
---
Arena ELO
49B
Parameters
131K
Context
Use this section to answer one simple question first: how much outside evidence do we have that this model performs well? Structured benchmark scores appear first, then official provider evidence, then live arena signal.
This model has official provider benchmark evidence, but not a normalized independent benchmark table yet.
A full normalized benchmark row is not available for this model yet.
Official benchmark evidence is already available below, so this release is still being tracked while the normalized table catches up.
Sign in to join the discussion
0
Downloads
0
Likes
Oct 2025
Released
These are recent benchmark or leaderboard claims from official provider sources. They are useful for freshness and context, but they are not treated the same as normalized independent benchmark rows.
Try NVIDIA NIM APIs
For voice agents and brand ambassadors. B200 + 21 248 2d Stepfun-ai Downloadable Free Endpoint step-3.7-flash A sparse MoE multimodal reasoning model good for enterprise, agentic and coding tasks. B200 + 5 1.21M 5d Moonshotai Downloadable Free Endpoint kimi-k2.6 1T multimodal MoE for long-horizon coding, agentic tool use, and image/video understanding.
View source