mlabonne / LFM2.5-230M-Chess

huggingface.co
Total runs: 1.7K
24-hour runs: 0
7-day runs: 113
30-day runs: 1.5K
Model's Last Updated: Septiembre 16 2026
text-generation

Introduction of LFM2.5-230M-Chess

Model Details of LFM2.5-230M-Chess

LFM2.5-230M-Chess

A 230M chess engine with a language model interface, fine-tuned from LFM2.5-230M-Base using Stockfish distillation.

Every possible chess move is a single token in the vocabulary. The model reads a position as a fixed 80-token prompt, predicts its own win probability, then predicts one move token. The host masks the move logits to the legal moves, so the model can never play an illegal move.

It plays at roughly 2004 Elo with a shallow depth-3 search on top, and roughly 1500 with the raw one-pass policy.

Demo: Play it in your browser with the ChessLFM Space .

Demo game

Usage
import json
from transformers import AutoModelForCausalLM

model = AutoModelForCausalLM.from_pretrained("mlabonne/LFM2.5-230M-Chess")
token_ids = json.load(open("token_ids.json"))

Each move takes two short forward passes:

  1. Build the 80-token prompt from the FEN. It encodes the 64 squares, side to move, castling rights, en passant file, halfmove clock, repetition count, and the last 8 plies.
  2. Argmax over the 64 value tokens gives the model evaluation, where <v:k> means a win probability in [k/64, (k+1)/64) .
  3. Append that value token and <|bestmove|> , mask the move logits to the legal moves, and take the argmax.

You can find the 2,106 added token IDs in token_ids.json (it starts at 64402).

Runs of mlabonne LFM2.5-230M-Chess on huggingface.co

1.7K
Total runs
0
24-hour runs
51
3-day runs
113
7-day runs
1.5K
30-day runs

More Information About LFM2.5-230M-Chess huggingface.co Model

More LFM2.5-230M-Chess license Visit here:

https://choosealicense.com/licenses/lfm1.0

LFM2.5-230M-Chess huggingface.co

LFM2.5-230M-Chess huggingface.co is an AI model on huggingface.co that provides LFM2.5-230M-Chess's model effect (), which can be used instantly with this mlabonne LFM2.5-230M-Chess model. huggingface.co supports a free trial of the LFM2.5-230M-Chess model, and also provides paid use of the LFM2.5-230M-Chess. Support call LFM2.5-230M-Chess model through api, including Node.js, Python, http.

LFM2.5-230M-Chess huggingface.co Url

https://huggingface.co/mlabonne/LFM2.5-230M-Chess

mlabonne LFM2.5-230M-Chess online free

LFM2.5-230M-Chess huggingface.co is an online trial and call api platform, which integrates LFM2.5-230M-Chess's modeling effects, including api services, and provides a free online trial of LFM2.5-230M-Chess, you can try LFM2.5-230M-Chess online for free by clicking the link below.

mlabonne LFM2.5-230M-Chess online free url in huggingface.co:

https://huggingface.co/mlabonne/LFM2.5-230M-Chess

LFM2.5-230M-Chess install

LFM2.5-230M-Chess is an open source model from GitHub that offers a free installation service, and any user can find LFM2.5-230M-Chess on GitHub to install. At the same time, huggingface.co provides the effect of LFM2.5-230M-Chess install, users can directly use LFM2.5-230M-Chess installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

LFM2.5-230M-Chess install url in huggingface.co:

https://huggingface.co/mlabonne/LFM2.5-230M-Chess

Url of LFM2.5-230M-Chess

LFM2.5-230M-Chess huggingface.co Url

Provider of LFM2.5-230M-Chess huggingface.co

mlabonne
ORGANIZATIONS

Other API from mlabonne

huggingface.co

Total runs: 156
Run Growth: -18
Growth Rate: -11.54%
Updated:Marzo 04 2024
huggingface.co

Total runs: 125
Run Growth: -42
Growth Rate: -32.56%
Updated:Marzo 04 2024
huggingface.co

Total runs: 62
Run Growth: -11
Growth Rate: -17.74%
Updated:Marzo 04 2024