whatmodel.sh / GPO-Llama-3-8B-Instruct-GPM-2B
GPO-Llama-3-8B-Instruct-GPM-2B
↗ share
+ contribute
GPO-Llama-3-8B-Instruct-GPM-2B is an open-weights language model from general-preference (GPM), released October 2024. Its published size is 8.03B parameters. Licence: apache-2.0.
Catalogue identity notes
Maker repo general-preference/GPO-Llama-3-8B-Instruct-GPM-2B @f6fae48: licence tag apache-2.0. 8,030,261,248 parameters from the Hugging Face safetensors metadata (dense, so active = total). Released 2024-10: month the maker's Hugging Face repo was created (createdAt 2024-10-08).
$ how it compares
The LLMs ranked either side of GPO-Llama-3-8B-Instruct-GPM-2B, out of 1132 with a score, then more from general-preference (GPM).
| # | model | score | where | ||
|---|---|---|---|---|---|
| 570 |  Arch-Agent-1.5B | 50±5 | local | ||
| 571 |  OpenCodeInterpreter-DS-33B | 50±1 | local | ||
| 572 | GPO-Llama-3-8B-Instruct-GPM-2B | 50±2 | local | ||
| 573 |  Llama 3.2 90B Vision Instruct | 50±1 | local | ||
| 574 |  DeepSeek-Coder 33B Instruct | 50±1 | local | ||
| more from general-preference (GPM) | |||||
| 611 | SPPO-Llama-3-8B-Instruct-GPM-2B | 49±2 | local | ||
$ what it's good at
chat & writing▁▁▁▄▆▅▆█▆▆▆▂ #278 of 466 * · 99% of its score · 1 test
Beats Ministral 3 14B Reasoning 2512, Ministral 3 14B, OpenThinker2 32B here, though they rank higher overall. Trails Flan-T5 XXL, Granite 4.0 H 1B, Granite 4.0 1B here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| AlpacaEval 2 length-controlled | 38.4% ✎ | ▁▂▂▃▆▆█▆▄▄▁▁ beats 95/122 | 99% |
$ where to run it
Other ways to run it
- Run it as published · needs ~16.06 GB · from huggingface.co/general-preference (official)
uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='general-preference/GPO-Llama-3-8B-Instruct-GPM-2B', torch_dtype='auto', device_map='auto')('Once upon a time', max_new_tokens=40)[0]['generated_text'])"
Official weights and documentation: huggingface.co/general-preference ↗
$ how the score was worked out · AlpacaEval 1.0 and 2.0 leader… 99%
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 |
|---|---|---|
 AlpacaEval 1.0 and 2.0 leaderboards (tatsu-lab/alpaca_eval repository) | 1 | 99% |
| prior · pull toward the middle | 1% | |
| weighted average | = 50.0 |
$ 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