[whatmodel.sh](../../README.md) / Llama 3.1 Nemotron 70B Instruct
# Llama 3.1 Nemotron 70B Instruct
NVIDIA · released 2024-10 · 70.6B params · llama3.1
**score 52.8** ±0.9 · #321 of 732 · from 19 benchmark results across 10 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks Llama 3.1 Nemotron 70B Instruct among every model on that task or test, worst to best.
task vs the others rank counts tests
coding ▁▁▁▃▅▇●▇▅▄▃▁ #278 of 469* 19% 2
chat & writing ▁▁▁▃▆▄▆●▆▇▇▃ #221 of 402 41% 6
math ▁▁▃▅●▆▄█▄▂▁▁ #312 of 431* 10% 2
coding · #278 of 469 · 19% of its score
Beats [Qwen3-VL 32B Instruct](qwen3-vl-32b-instruct.md), [Solar Open 100B (102B-A12B)](solar-open-100b.md), [GLM-4.7-Flash](glm-4-7-flash.md) here, though they rank higher overall. Trails [DeepSeek LLM 67B Chat](deepseek-llm-67b-chat.md), [Mistral NeMo Instruct (2407)](mistral-nemo-instruct-2407.md), [QwQ-32B-Preview](qwq-32b-preview.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
Aider code editing (133 Exercism … 54.9% ▁·▁▂▄▁●▄█▆▂▂ beats 21/62 medium · avg 51% 15% sharp · at its level
WebApp1K-React pass@1 (onekq-ai) 22.4% ▃▁●█▁▄▅▃▄▇▁▄ beats 3/31 medium · avg 40% 3.8% sharp · at its level · noisy
chat & writing · #221 of 402 · 41% of its score
Beats [Gemma 2 9B IT SimPO](gemma-2-9b-simpo.md), [Granite 4.2 8B](granite-4-2-8b.md), [Mistral Small 4 (119B-A6.5B)](mistral-small-4.md) here, though they rank higher overall. Trails [LFM2.5 1.2B Instruct](lfm2-5-1-2b-instruct.md), [Granite 4.0 H Micro 3B](granite-4-0-h-micro.md), [Gemma 3n E4B](gemma-3n-e4b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
Arena-Hard (v0.1, GPT-4-1106 judg… 84.9% ▂▁▁▅█▅▄▃▃▅●▁ beats 56/60 medium · avg 47% 20% sharp · at its level
LMArena (text) 1298.7 ▁▁▁▂▃▃▃●▆▇█▃ beats 127/385 avg 1260 8.1% blunt
Arena-Hard v2.0 hard prompts (sty… 10.3% ▂·●▅▅·▃█▆▂▂▂ beats 2/22 very hard · avg 12% 5.9% sharp
AlpacaEval 2 length-controlled 66.1% ▁▁▃▄▆██▆▃●▁▁ beats 146/155 medium · avg 43% 3.6% sharp · at its level · small run
IFEval prompt-level loose 79.9% ▁▂▁▂▁▃▁●█▂▂▁ beats 11/25 easy · avg 78% 2.0% small run
Arena-Hard v2.0 creative writing … 26.9% ▆▂▂●▂▂▃█▂▃·▂ beats 8/22 hard · avg 19% 1.6% sharp · noisy
math · #312 of 431 · 10% of its score
Beats [Molmo 2 8B](molmo-2-8b.md) here, though they rank higher overall. Trails [DeepSeek R1 Distill Qwen 1.5B](deepseek-r1-distill-qwen-1-5b.md), [Gemma 7B (base)](gemma-7b-base.md), [Mistral 7B v0.1 (base)](mistral-7b-v0-1.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
GSM8K 84.7% ▂▂▂▂▃▄▅▇█●▆▄ beats 151/206 easy · avg 84% 5.8% sharp · small run
MATH (subset unspecified) 55.0% ▂▄▄▃●▇▅▆▄▃▄▁ beats 31/80 medium · avg 54% 4.7% at its level · small run
other tests (9)
test result vs the others in its run difficulty counts why
BIG-Bench Hard 78.5% ▁▁▁▁▂▄▅▆█●▃▁ beats 133/155 easy · avg 75% 6.6% sharp · at its level · small run
MMLU 83.8% ▁▁▁▁▁▂▂▅██●▄ beats 182/214 easy · avg 77% 5.3% small run
BRIDGE clinical text, zero-shot a… 32.8% ▁▁▁▁▁▂▂▃●▆█▃ beats 38/97 hard · avg 31% 4.7% blunt · at its level
LiveBench 2024-08-31 43.1% ●█·▄·▄▄·▄··▄ beats 0/7 medium · avg 40% 4.5% at its level
EvalPerf correctness (118 tasks, … 61.8% ▂·▂●▂▆▅▆▆▆▃█ beats 2/29 easy · avg 74% 4.2% typical
PopQA 36.4% ▁▂▁▃▃█▂█▆●▁▆ beats 33/40 hard · avg 34% 1.9% at its level · small run
EuroEval Life in the UK (English … 93.5% ▁▁▁▁▁▁▁▂▄█▄● beats 131/146 easy · avg 82% 0.9% noisy
DROP 68.8% ▃▅█▅·▃·▆●▂▁▃ beats 34/44 medium · avg 62% 0.7% blunt · at its level · small run
TruthfulQA MC2 62.6% █▇▇▄▂▄▁●▅▄▁▁ beats 53/79 medium · avg 59% 0.2% blunt · at its level · small run
source results share of the score
Tülu 3: Pushing Frontiers in Open Langu… 9 31% ▇▇▇▇
Arena-Hard-Auto v0.1 leaderboard 1 20% ▇▇
Aider code editing benchmark runs 1 15% ▇▇
LMArena leaderboard dataset 1 8% ▇
Arena-Hard-Auto v2.0-Preview leaderboard 2 7% ▇
BRIDGE Medical Leaderboard 1 5% ▇
LiveBench model judgments 1 4% ▇
EvalPlus / EvalPerf published evaluatio… 1 4% ▇
WebApp1K models leaderboard 1 4% ▇
EuroEval leaderboards 1 1% ▇
prior: pull toward the middle 0% ▇
weighted average = 52.8
Licences: Tülu 3: Pushing Frontiers in Open Language Model Post-Training ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Arena-Hard-Auto v0.1 leaderboard ([Apache-2.0](../../licenses/arena-hard-auto-v0-1-Apache-2.0.txt)) · Aider code editing benchmark runs ([Apache-2.0](../../licenses/aider-polyglot-runs-Apache-2.0.txt)) · LMArena leaderboard dataset ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Arena-Hard-Auto v2.0-Preview leaderboard ([Apache-2.0](../../licenses/arena-hard-auto-Apache-2.0.txt)) · BRIDGE Medical Leaderboard ([MIT](../../licenses/bridge-medical-leaderboard-MIT.txt)) · LiveBench model judgments ([Apache-2.0 (stated in the LiveBench datasheet)](https://www.apache.org/licenses/LICENSE-2.0)) · EvalPlus / EvalPerf published evaluation runs ([Apache-2.0](../../licenses/evalplus-published-runs-Apache-2.0.txt)) · WebApp1K models leaderboard ([MIT](../../licenses/webapp1k-leaderboard-MIT.txt)) · EuroEval leaderboards ([MIT](https://github.com/EuroEval/leaderboards/blob/1a77f1d2f8ba1ba05ab73287fb090d76bf41b2e0/LICENSE))
## How to run it
### llama.cpp (recommended)
1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md)
```sh
# macOS / Linux
curl -LsSf https://llama.app/install.sh | sh
# Windows
irm https://llama.app/install.ps1 | iex
```
2. Run Llama 3.1 Nemotron 70B Instruct (Q4_K_M, needs ~42.5 GB, from [huggingface.co/lmstudio-community](https://huggingface.co/lmstudio-community/Llama-3.1-Nemotron-70B-Instruct-HF-GGUF/tree/672e9a6fa26d4b37bc118993d49622b174b8d4b0))
```sh
llama cli -hf lmstudio-community/Llama-3.1-Nemotron-70B-Instruct-HF-GGUF:Q4_K_M
```
3. Or start a local server and open the address it prints (http://localhost:9931)
```sh
llama serve -hf lmstudio-community/Llama-3.1-Nemotron-70B-Instruct-HF-GGUF:Q4_K_M
```
Other versions: Q8_0 (~75 GB) `llama cli -hf lmstudio-community/Llama-3.1-Nemotron-70B-Instruct-HF-GGUF:Q8_0`
### LM Studio
1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download)
2. Run Llama 3.1 Nemotron 70B Instruct (Q4_K_M, needs ~42.5 GB, from [huggingface.co/lmstudio-community](https://huggingface.co/lmstudio-community/Llama-3.1-Nemotron-70B-Instruct-HF-GGUF/tree/672e9a6fa26d4b37bc118993d49622b174b8d4b0))
```sh
lms get https://huggingface.co/lmstudio-community/Llama-3.1-Nemotron-70B-Instruct-HF-GGUF@Q4_K_M -y
```
3. Then run lms ls to see its name, and lms chat followed by that name
Other versions: Q8_0 (~75 GB) `lms get https://huggingface.co/lmstudio-community/Llama-3.1-Nemotron-70B-Instruct-HF-GGUF@Q8_0 -y` · 4bit (~39.7 GB, Mac only) `lms get https://huggingface.co/mlx-community/nvidia_Llama-3.1-Nemotron-70B-Instruct-HF_4bit -y` · 8bit (~75 GB, Mac only) `lms get https://huggingface.co/mlx-community/Llama-3.1-Nemotron-70B-Instruct-HF-8bit -y`
### Ollama
1. Install Ollama. [Download it (macOS / Windows)](https://ollama.com/download)
```sh
# Linux
curl -fsSL https://ollama.com/install.sh | sh
```
2. Run Llama 3.1 Nemotron 70B Instruct (Q4_K_M, needs ~42.86 GB, from [ollama.com](https://ollama.com/library/nemotron/tags))
```sh
ollama run nemotron:70b-instruct-q4_K_M
```
Other versions: Q8_0 (~85.22 GB) `ollama run nemotron:70b-instruct-q8_0`
### MLX (Mac)
1. Install MLX (Mac). Apple silicon Macs only, text only.
```sh
# with uv
uv tool install mlx-lm
# or
pip install mlx-lm
```
2. Run Llama 3.1 Nemotron 70B Instruct (4bit, needs ~39.7 GB, Mac only, from [huggingface.co/mlx-community](https://huggingface.co/mlx-community/nvidia_Llama-3.1-Nemotron-70B-Instruct-HF_4bit/tree/b40b6355521e949ad45c71d0c7db8e6cd1a54cdd))
```sh
mlx_lm.chat --model mlx-community/nvidia_Llama-3.1-Nemotron-70B-Instruct-HF_4bit
```
3. Or start an OpenAI-compatible server at http://localhost:8080/v1
```sh
mlx_lm.server --model mlx-community/nvidia_Llama-3.1-Nemotron-70B-Instruct-HF_4bit
```
Other versions: 8bit (~75 GB, Mac only) `mlx_lm.chat --model mlx-community/Llama-3.1-Nemotron-70B-Instruct-HF-8bit`
Model repository and documentation: [huggingface.co/nvidia](https://huggingface.co/nvidia/Llama-3.1-Nemotron-70B-Instruct-HF)
[interactive version](https://whatmodel.sh/model/llama-3-1-nemotron-70b) · [JSON](https://whatmodel.sh/api?format=json&all=1)