whatmodel.sh / Llama 3 70B (base)
Llama 3 70B (base)
↗ share
+ contribute
Llama 3 70B (base) is an open-weights language model from Meta, released April 2024. Its published size is 70.55B parameters. Licence: llama3.
Catalogue identity notes
Maker repo meta-llama/Meta-Llama-3-70B @c824948: licence tag llama3. 70,553,706,496 parameters from the Hugging Face safetensors metadata (dense, so active = total). Released 2024-04: month the maker's Hugging Face repo was created (createdAt 2024-04-17). Pretrained base checkpoint (not an instruct/chat version). Gated repository (licence acceptance required on Hugging Face).
$ how it compares
The LLMs ranked either side of Llama 3 70B (base), out of 828 with a score, then more from Meta.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 497 | Code Llama 70B Instruct | 46±1 | local | ||
| 498 |  Hermes 2 Pro Llama 3 8B | 46±3 | local | ||
| 499 | Llama 3 70B (base) | 46±2 | local | ||
| 500 |  Granite 4.0 Micro 3B | 46±1 | local | ||
| 501 |  Command R (08-2024) | 46±1 | api $0.26/M +local | ||
| more from Meta | |||||
| 24 | Muse Spark 1.3 | 77±1 | api $2.00/M | ||
| 40 | Muse Spark 1.2 | 75±1 | api $2.00/M | ||
| 43 | Muse Spark 1.1 | 75±1 | api $2.00/M | ||
$ what it's good at
other tests4 not in a task · 98% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| BigCodeBench (Complete) | 43.3% ✎ | ▂▁▁▄▂▅▇█▆▂▂▂ beats 36/58 | 55% |
| Polish EQ-Bench v2 (SpeakLeash) | 46.1% ✎ | ▁··▁▁▂▂▆▅▆█▆ beats 18/70 | 27% |
| Open PT-LLM Leaderboard average (9 Portuguese tasks) | 77.9% ✎ | ▃▁▁▁▁▁▁▁▂█▅▁ beats 242/276 | 10% |
| EuroEval Life in the UK (English knowledge MC), test split, 5-shot, accuracy | 91.8% ✎ | ▁▁▁▁▁▁▁▂▄█▄▃ beats 123/146 | 6.0% |
$ where to run it
LM Studio · recommended
- Install LM Studio · Open the app once before using the lms command.
- Run Llama 3 70B (base) · needs ~39.7 GB, Mac only · from huggingface.co/mlx-community
- Or download it in the terminal
lms get https://huggingface.co/mlx-community/Meta-Llama-3-70B-4bit -y - Then run lms ls to see its name, and lms chat followed by that name
MLX (Mac)
- Install MLX (Mac) · Apple silicon Macs only, text only.with uv
uv tool install mlx-lmorpip install mlx-lm - Run Llama 3 70B (base) · needs ~39.7 GB, Mac only · from huggingface.co/mlx-communitychat in the terminal
mlx_lm.chat --model mlx-community/Meta-Llama-3-70B-4bit - Or start an OpenAI-compatible server at http://localhost:8080/v1
mlx_lm.server --model mlx-community/Meta-Llama-3-70B-4bit
Other ways to run it
- Run it as published · needs ~141.11 GB · from huggingface.co/meta-llama
uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='meta-llama/Meta-Llama-3-70B', torch_dtype='auto', device_map='auto')('Once upon a time', max_new_tokens=40)[0]['generated_text'])"
Weights and documentation: huggingface.co/meta-llama ↗
$ how the score was worked out · BigCodeBench paper (Zhuo et a… 55%, SpeakLeash Polish EQ-Bench le… 27%, Open Portuguese LLM Leaderboa… 10% and 1 more
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 |
|---|---|---|
 BigCodeBench paper (Zhuo et al., ICLR 2025), arXiv 2406.15877v4 | 1 | 55% |
 SpeakLeash Polish EQ-Bench leaderboard | 1 | 27% |
| Open Portuguese LLM Leaderboard (archived) | 1 | 10% |
 EuroEval leaderboards (results collection), English generative leaderboard | 1 | 6% |
| prior · pull toward the middle | 2% | |
| weighted average | = 46.1 |
$ 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