🏆 #6,974 overall#1372 of 1,375 in Language Models🔥 active this week

JaydenTeoh /Soft-MTP

Soft-MTP: Multi-Token Prediction Done Right. KL gradient recovery for faster speculative decoding and better language model training.

$ git clone https://github.com/JaydenTeoh/Soft-MTP.git
GitHub social preview for JaydenTeoh/Soft-MTP
Stars
100
100
This week
—
Forks
13
13
Language
Python
License
MIT
Created
Oct 27, 2026
0.0 years old
Last push
Nov 12, 2026
🔥 this week

Categories

GitHub topics

More in Language Models

Language Models

JaydenTeoh/NextLat

▲ +4 ★ this week

Official codebase for "Next-Latent Prediction Transformers Learn Compact World Models"

PythonMITupdated Jun 12, 2026
GitHub ↗★ 216⑂ 16
Language Models

huggingface/tokenizers🔥 active

▲ +36 ★ this week

💥 Fast State-of-the-Art Tokenizers optimized for Research and Production

🧠 Natural Language Processing
RustApache-2.0updated Nov 12, 2026
GitHub ↗★ 11.3K⑂ 1.3K
Language Models

EleutherAI/gpt-neox🔥 active

▲ +4 ★ this week

An implementation of model parallel autoregressive transformers on GPUs, based on the Megatron and DeepSpeed libraries

PythonApache-2.0updated Nov 7, 2026
GitHub ↗★ 7.5K⑂ 1.1K
Language Models

ai-forever/ru-gpts

▲ +1 ★ this week

Russian GPT3 models.

PythonApache-2.0updated Dec 12, 2022
GitHub ↗★ 2.1K⑂ 431
Language Models

elicit/machine-learning-list

▲ +1 ★ this week

A curriculum for learning about foundation models, from scratch to the frontier

Otherno licenseupdated Nov 27, 2025
GitHub ↗★ 1.5K⑂ 126
Language Models

SKTBrain/KoBERT

▲ +1 ★ this week

Korean BERT pre-trained cased (KoBERT)

🧠 Natural Language Processing
PythonApache-2.0updated Jun 14, 2025
GitHub ↗★ 1.4K⑂ 374

Data from GitHub · snapshot Nov 13, 2026