Releases·Model Release
mlabonne releases LFM2.5-230M-Chess, a 230M-parameter chess model with an online demo on Hugging Face
mlabonne releases LFM2.5-230M-Chess model and online demo on Hugging Face
70 engagement1 sourceWed, Sep 16, 2026, 04:53 AM
AI Overview
- Hugging Face hosts mlabonne's LFM2.5-230M-Chess, a 230M-parameter model built to play chess.
- An online demo, the ChessLFM Hugging Face Space, lets users test the model's chess play directly in a browser.
- The model is mlabonne's fine-tune of the LFM2.5 architecture scaled down to 230M parameters for chess.
Related
Products·launched
Playable chess neural network with five training snapshots as browser opponents
A website where you play against snapshots of a single chess neural network taken during a 30-minute training run. All five snapshots run in the browser as opponents, can be compared side by side on any position, or pitted against each other. It demonstrates how a chess model improves as it learns, for casual players and the curious.
A website where you play against snapshots of a single chess neural network taken during a 30-minute training run. All five snapshots run in the browser as opponents, can be compared side by side on any position, or pitted against each other. It demonstrates how a chess model improves as it learns, for casual players and the curious.
1 engagement1 sourceredditSat, Sep 5, 2026, 09:31 PM
Developments·Paper
Chessformer_lens demo shows ablating one attention head breaks Morphy's queen sacrifice
A Reddit post demonstrates that ablating a single attention head in a chess transformer causes the model to stop finding Morphy's queen sacrifice, highlighting the critical role of individual heads. The demo includes GIFs and notebooks for replication.
A Reddit post demonstrates that ablating a single attention head in a chess transformer causes the model to stop finding Morphy's queen sacrifice, highlighting the critical role of individual heads. The demo includes GIFs and notebooks for replication.
16 engagement1 sourceredditThu, Aug 13, 2026, 12:29 AM
Releases·Model Release
Liquid AI releases LFM 2.5 QAD, a 2.6B quantized model
Liquid AI released LFM 2.5 QAD, a quantized version of its Liquid Foundation Model, with a 2.6B parameter GGUF file available on Hugging Face. The release was announced via a tweet from Liquid AI's official account.
Liquid AI released LFM 2.5 QAD, a quantized version of its Liquid Foundation Model, with a 2.6B parameter GGUF file available on Hugging Face. The release was announced via a tweet from Liquid AI's official account.
140 engagement1 sourceredditWed, Aug 19, 2026, 04:59 PM
Products·launched
Online chess website where AI coding agents play each other
RoboChess is a website where AI coding agents play chess against each other. Users can paste a link into Claude Code to have their agent join a game, and watch the games live. It provides a fun way to use extra tokens and showcases agent capabilities.
RoboChess is a website where AI coding agents play chess against each other. Users can paste a link into Claude Code to have their agent join a game, and watch the games live. It provides a fun way to use extra tokens and showcases agent capabilities.
8 engagement1 sourceredditThu, Jul 16, 2026, 09:49 PM
Products·launched
BYOK playground where LLMs must output legal moves in Chess, Go, Xiangqi, Gomoku, and Othello
A bring-your-own-key playground for playing board games against LLMs across five games. The model must emit a parseable move in text, a rules engine applies it, and illegal or unparseable output counts toward losing. It targets developers who want to benchmark LLMs without Stockfish wrappers or free-form rambling.
A bring-your-own-key playground for playing board games against LLMs across five games. The model must emit a parseable move in text, a rules engine applies it, and illegal or unparseable output counts toward losing. It targets developers who want to benchmark LLMs without Stockfish wrappers or free-form rambling.
3 engagement1 sourceredditMon, Sep 7, 2026, 04:51 PM