whatmodel.sh / Nemotron-Cascade-8B

Nemotron-Cascade-8B

+ contribute

Nemotron-Cascade-8B is an open-weights language model from NVIDIA, released December 2025. Its published size is 8.19B parameters. Licence: nvidia-open-model-license.

Catalogue identity notes

NVIDIA's own model, repo nvidia/Nemotron-Cascade-8B @276a779 (licence nvidia-open-model-license). Unified model (thinking and instruct modes), 'post-trained from the Qwen3-8B-Base model'. Printed as 'Nemotron-Cascade-8B' in the Nemotron-Cascade report (arXiv 2512.13607v2) Table 9; listed so the report's NVIDIA rows join the existing LiveCodeBench Pro 25Q2 runs.

Model information sources

Release date ↗ · Weights licence ↗

safetensors metadata of nvidia/Nemotron-Cascade-8B @276a779: 8,190,735,360 parameters (dense).

58.6±1.0#211 of 719 LLMsfrom 12 benchmark results across 1 source

$ how it compares

The LLMs ranked either side of Nemotron-Cascade-8B, out of 828 with a score, then more from NVIDIA.

#	model	score	where	RAM	released
251	EXAONE 4.0 32B LG AI Research	59±1	local	~20GB	2025-07
252	Phi-4-reasoning-plus 14B Microsoft	59±1	local	~10GB	2025-04
253	Nemotron-Cascade-8B NVIDIA	59±1	local	~6.3GB	2025-12
254	Qwen3 14B Alibaba	59±1	api $0.61/M +local	~10GB	2025-04
255	Mercury 2 Inception	59±2	comparison only
	more from NVIDIA
98	Nemotron 3 Ultra 550B-A55B NVIDIA	69±1	api $1.35/M +local	~89GB	2026-06
155	Nemotron Cascade 2 30B-A3B NVIDIA	66±1	local	~20GB	2026-03
168	Nemotron 3 Super 120B-A12B NVIDIA	65±1	api $0.28/M +local	~73GB	2026-03

see all LLMs · find the ones that fit your machine

$ what it's good at

agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #194 of 300 * · 7.4% of its score · 2 tests

Beats o1 (2024-12-17), o3-mini, DeepSeek R1 (Jan '25) here, though they rank higher overall. Trails SmolLM3 3B, Jamba 1.5 Mini, Mellum2 12B-A2.5B Instruct here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
SWE-bench Verified lab	37.2% ✎	▁▁▁▁▁▁▃▄█▆▁▁ beats 26/131	hard · avg 19%	▇▇ 5.9%	sharp · at its level · small run
BFCL v3 lab	64.4% ✎	▁▁▁▁▁▃▅█▆▅▄▁ beats 122/170	medium · avg 58%	▇ 1.4%	blunt · at its level · small run

coding▁▁▁▃▆██▇▆▄▃▁ #192 of 458 · 53% of its score · 4 tests

Beats Gemini 2.5 Flash, Qwen3.5 4B, Qwen3 30B-A3B Instruct 2507 here, though they rank higher overall. Trails QwQ-32B-Preview, DeepSeek-Coder-V2 Instruct (0724), DeepSeek-V2.5 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
LiveCodeBench Pro 2025Q2 (Medium) lab	6.4% ✎	▃▃····▃·█·▃▃ beats 5/7	very hard · avg 2.4%	▇▇▇▇▇▇▇▇▇▇▇ 28%	sharp · too hard for it
LiveCodeBench Pro 2025Q2 (Easy) lab	65.7% ✎	▄·▄····▄▄▄█▄ beats 5/7	medium · avg 45%	▇▇▇▇▇ 13%	at its level
SWE-bench Verified lab	37.2% ✎	▁▁▁▁▁▁▃▄█▆▁▁ beats 26/131	hard · avg 19%	▇▇ 5.9%	sharp · at its level · small run
LiveCodeBench v6 (Aug 2024–May 2025) lab	71.1% ✎	▄▄···▄·▄▄█·▄ beats 5/7	medium · avg 57%	▇▇ 5.8%	at its level

chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #108 of 393 * · 1.9% of its score · 1 test

Beats LongCat 2.0 (1.6T-A48B), GPT-5.4 nano, Step 3.5 Flash (196B-A11B) here, though they rank higher overall.

test	result	vs the others in its run	difficulty	counts	why
IFEval prompt-level strict lab	90.2% ✎	▁▁··▂··▂▇█▃▃ beats 24/26	easy · avg 77%	▇ 1.9%	small run

math▁▂▃▅▆▅▄█▄▁▁▁ #131 of 398 * · 18% of its score · 2 tests

Beats Claude Sonnet 4.6, MiniMax-M3, MiMo-V2.5 (310B-A15B) here, though they rank higher overall. Trails Olmo 3.1 32B Think here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
AIME 2025 lab	80.1% ✎	▁·▁▁▁▂▅▆█▇▅▁ beats 85/133	medium · avg 55%	▇▇▇▇ 10%	sharp · small run
AIME 2024 lab	89.5% ✎	▁▁▁▂▁▃▁▅███▃ beats 70/85	easy · avg 68%	▇▇▇▇ 8.9%	sharp · small run

other tests4 not in a task · 25% of its score

test	result	vs the others in its run	difficulty	counts	why
GPQA Diamond lab	66.5% ✎	▁▁▁▁▂▄▅▅███▄ beats 138/310	hard · avg 46%	▇▇▇▇ 11%	sharp · at its level · small run
MMLU lab	83.7% ✎	▁▁▁▁▁▂▃▅█▃▆▃ beats 150/183	medium · avg 73%	▇▇ 5.9%	small run
MMLU-Pro lab	75.7% ✎	▁▁·▁▁▁▃▂▄▅█▁ beats 104/188	medium · avg 60%	▇▇ 5.0%	at its level · small run
IFBench (aggregation unspecified) lab	40.8% ✎	▃▇▂▄▅▅▂▄▄▆█▇ beats 27/80	hard · avg 31%	▇ 3.4%	at its level · small run

$ 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 | sh
Windows
irm https://llama.app/install.ps1 | iex
other ways to install ↗
Run Nemotron-Cascade-8B · needs ~5.03 GB · from huggingface.co/bartowski
chat in the terminal
llama cli -hf bartowski/nvidia_Nemotron-Cascade-8B-GGUF:Q4_K_M
Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf bartowski/nvidia_Nemotron-Cascade-8B-GGUF:Q4_K_M

LM Studio

version

Install LM Studio · Open the app once before using the lms command.
download it (macOS / Windows / Linux) ↗
Run Nemotron-Cascade-8B · needs ~5.03 GB · from huggingface.co/bartowski
open in LM Studio ↗
Or download it in the terminal
lms get https://huggingface.co/bartowski/nvidia_Nemotron-Cascade-8B-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) ↗
Linux
curl -fsSL https://ollama.com/install.sh | sh
Run Nemotron-Cascade-8B · needs ~5.03 GB · from huggingface.co/bartowski
chat in the terminal
ollama run hf.co/bartowski/nvidia_Nemotron-Cascade-8B-GGUF:Q4_K_M

MLX (Mac)

Install MLX (Mac) · Apple silicon Macs only, text only.
with uv
uv tool install mlx-lm
or
pip install mlx-lm
Run Nemotron-Cascade-8B · needs ~4.6 GB, Mac only · from huggingface.co/introvoyz041
chat in the terminal
mlx_lm.chat --model introvoyz041/Nemotron-Cascade-8B-mlx-4Bit
Or start an OpenAI-compatible server at http://localhost:8080/v1
mlx_lm.server --model introvoyz041/Nemotron-Cascade-8B-mlx-4Bit

Official weights and documentation: huggingface.co/nvidia ↗

$ how the score was worked out · NVIDIA — Nemotron-Cascade: Sc… 100%

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	licence
NVIDIA — Nemotron-Cascade: Scaling Cascaded Reinforcement Learning for General-Purpose Reasoning Models · lab-reported	12	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 100%	CC BY 4.0
prior · pull toward the middle		▇ 0%
weighted average		= 58.6

$ improve this page

Update model information Add a place to run it Add a test result Propose a benchmark Correct a benchmark score Edit data on GitHub

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 2026-10-11 02:20 UTC · sources and licences · json