whatmodel.sh / Llama 3.3 Nemotron Super 49B v1
Llama 3.3 Nemotron Super 49B v1
↗ share
+ contribute
Llama 3.3 Nemotron Super 49B v1 is an open-weights language model from NVIDIA, released March 2025. Its published size is 49.9B parameters. Licence: nvidia-open-model-license.
Catalogue identity notes
Comparison-only. Arena entry 'llama-3.3-nemotron-49b-super-v1' (text). LMArena's leaderboard table (arena-leaderboard space @6f2eeeb) labels 'llama-3.3-nemotron-49b-super-v1' as 'Llama-3.3-Nemotron-Super-49B-v1' (Nvidia, Nvidia), linking https://huggingface.co/nvidia/Llama-3_3-Nemotron-Super-49B-v1. Maker repo nvidia/Llama-3_3-Nemotron-Super-49B-v1 @387156d: licence tag other (nvidia-open-model-license). 49,867,145,216 parameters from the Hugging Face safetensors metadata (dense, so active = total). Released 2025-03: month the maker's Hugging Face repo was created (createdAt 2025-03-16). v1 (March 2025), distinct from the catalogued v1.5 (llama-3-3-nemotron-super-49b-v1-5).
$ how it compares
The LLMs ranked either side of Llama 3.3 Nemotron Super 49B v1, out of 828 with a score, then more from NVIDIA.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 256 |  DiffusionGemma 26B-A4B | 59±1 | comparison only | ||
| 257 |  gpt-oss-20b | 58±1 | api $0.13/M +local | ||
| 258 | Llama 3.3 Nemotron Super 49B v1 | 58±2 | local | ||
| 259 |  Olmo 3.1 32B Think | 58±1 | local | ||
| 260 |  DeepSeek R1 Distill Llama 70B | 58±1 | api $0.80/M +local | ||
| more from NVIDIA | |||||
| 98 | Nemotron 3 Ultra 550B-A55B | 69±1 | api $1.35/M +local | ||
| 155 | Nemotron Cascade 2 30B-A3B | 66±1 | local | ||
| 168 | Nemotron 3 Super 120B-A12B | 65±1 | api $0.28/M +local | ||
$ what it's good at
agent tasksnot ranked on this task yet · 1 test
| test | result | vs the others in its run | counts |
|---|---|---|---|
| BFCL v2 Live lab | 73.7% ✎ | █··········█ beats 0/1 | — |
codingnot ranked on this task yet · 1 test
| test | result | vs the others in its run | counts |
|---|---|---|---|
| LiveCodeBench 2024-08 to 2025-02 (version not stated) lab | 45.5% ✎ | █··········█ beats 0/1 | — |
chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #180 of 393 · 37% of its score · 3 tests
Beats Seed-OSS 36B Instruct, Nemotron-Cascade-14B-Thinking, Granite 4.2 8B here, though they rank higher overall. Trails LFM2.5 1.2B Instruct, LFM2.5 1.2B Thinking, Gemma 3 12B here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| LMArena (text) | 1327.8 ✎ | ▁▁▁▂▃▃▃▆▆▇█▃ beats 166/379 | 23% |
| Arena-Hard (version/judge unspecified) lab | 88.3% ✎ | ▃▃▂▅▃▆▅█▅▅▂▃ beats 30/30 | 8.2% |
| IFEval (aggregation unspecified) lab | 89.2% ✎ | ▁▁▃▁▂▄▄▆█▃▂▁ beats 108/141 | 5.0% |
math█··········█ #1 of 2 · 43% of its score · 3 tests
other tests1 not in a task · 19% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| GPQA Diamond lab | 66.7% ✎ | ▁▁▁▁▂▄▅▅███▄ beats 140/310 | 19% |
$ where to run it
llama.cpp · recommended
version
- Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.macOS / Linux
curl -LsSf https://llama.app/install.sh | shWindowsirm https://llama.app/install.ps1 | iex - Run Llama 3.3 Nemotron Super 49B v1 · needs ~30.22 GB · from huggingface.co/unslothchat in the terminal
llama cli -hf unsloth/Llama-3_3-Nemotron-Super-49B-v1-GGUF:Q4_K_M - Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf unsloth/Llama-3_3-Nemotron-Super-49B-v1-GGUF:Q4_K_M
LM Studio
version
- Install LM Studio · Open the app once before using the lms command.
- Run Llama 3.3 Nemotron Super 49B v1 · needs ~30.22 GB · from huggingface.co/unsloth
- Or download it in the terminal
lms get https://huggingface.co/unsloth/Llama-3_3-Nemotron-Super-49B-v1-GGUF@Q4_K_M -y - Then run lms ls to see its name, and lms chat followed by that name
Ollama
version
- Install Ollama
download it (macOS / Windows) ↗
Linuxcurl -fsSL https://ollama.com/install.sh | sh - Run Llama 3.3 Nemotron Super 49B v1 · needs ~30.22 GB · from huggingface.co/unslothchat in the terminal
ollama run hf.co/unsloth/Llama-3_3-Nemotron-Super-49B-v1-GGUF:Q4_K_M
MLX (Mac)
version
- Install MLX (Mac) · Apple silicon Macs only, text only.with uv
uv tool install mlx-lmorpip install mlx-lm - Run Llama 3.3 Nemotron Super 49B v1 · needs ~28.1 GB, Mac only · from huggingface.co/mlx-communitychat in the terminal
mlx_lm.chat --model mlx-community/Llama-3_3-Nemotron-Super-49B-v1-mlx-4bit - Or start an OpenAI-compatible server at http://localhost:8080/v1
mlx_lm.server --model mlx-community/Llama-3_3-Nemotron-Super-49B-v1-mlx-4bit
Official weights and documentation: huggingface.co/nvidia ↗
$ how the score was worked out · NVIDIA — Llama-Nemotron: Effi… 75%, LMArena leaderboard dataset 23%
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
 NVIDIA — Llama-Nemotron: Efficient Reasoning Models · lab-reported | 6 | 75% |
 LMArena leaderboard dataset | 1 | 23% |
| prior · pull toward the middle | 1% | |
| weighted average | = 58.3 |
not counted (2)
| NVIDIA — Llama-Nemotron: Efficient Reasoning Models | BFCL v2 Live [correct] | 73.7% | fewer than three models have this test yet, or this source ran it for this model alone |
| NVIDIA — Llama-Nemotron: Efficient Reasoning Models | LiveCodeBench 2024-08 to 2025-02 (version not stated) [correct] | 45.5% | fewer than three models have this test yet, or this source ran it for this model alone |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json