Agent frameworks & infrastructure
Models, fine-tuning & training
2,504 agents · most starred first
Models, fine-tuning & training, page 5 of 26
Build your own offline AI from any documents. Free. No coding. LoRA fine-tuning + RAG + GGUF export.
FlashREINFORCE: Critic-Free, Single-Rollout, Asynchronous RL for Agentic Language Models
WangChanGLM 🐘 - The Multilingual Instruction-Following Model
Official code for paper "SPA-RL: Reinforcing LLM Agent via Stepwise Progress Attribution"
AgentHER: Hindsight Experience Replay for LLM Agents
[EMNLP 2026 main] Agent-G2 : Gaussian Guidance for Agentic Reinforcement Learning
A simple example of how to implement vector based DQN using PyTorch and a ML-Agents environment
transformers pytorch safetensors llama text-generation license:other
More Methods, Better Agents? A Reality Check on Self-Distillation in Agentic Reinforcement Learning
Implementation of Agent Attention in Pytorch
High-performance .NET BPE tokenizer — up to 618 MiB/s, competitive with Rust. Zero-allocation counting, multilingual cache, o200k/cl100k/r50k/p50k encodings + HuggingFace tokenizer.json support.
transformers pytorch gpt_neox text-generation en dataset:togethercomputer/RedPajama-Data-1T
transformers safetensors qwen2 text-generation conversational en
Black-box LLM distillation for any verifiable task, with text or image inputs: teacher API -> rejection sampling -> SFT -> eval -> GRPO
From dataset to trained LoRA to finished image or video. One browser workspace for your models, datasets, training, and generation. LoRA Pilot supports 50+ training model groups, with ComfyUI, Kohya, AI Toolkit, and Diffusion Pipe included. Run on your own GPU or RunPod, and connect AI agents through MCP.
transformers safetensors qwen2 text-generation code qwen-coder
Agentic Kernel Optimization — advanced & eXtensible: a closed-loop, campaign-based multi-agent system for optimizing GPU kernels (benchmark-swappable; default flashinfer-bench).
transformers safetensors stablelm text-generation causal-lm conversational
Dialog State Tracking Challenge 2 & 3 Data
A simple stochastic OpenAI environment for training RL agents
gguf code codeqwen chat qwen qwen-coder
Code for Agentic ESOpt: Fine-Tuning Long-Horizon LLM Agents with Minimal GPU Requirements
transformers safetensors qwen2 text-generation chat conversational
The official codebase for our paper, FLEX: Continuous Agent Evolution via Forward Learning from Experience.
Swift BPE Encoder/Decoder for OpenAI GPT Models. A programmatic interface for tokenizing text for OpenAI ChatGPT API.
A Practitioner's Guide to M(eow)ti Turn Agentic ReinfOrcement learning
A large collection of mostly reinforcement learning agents.
Collection of Deep Reinforcement Learning Algorithms implemented in PyTorch.
Agent Based Modeling and Reinforcement Learning
transformers safetensors gemma4_assistant text-generation any-to-any arxiv:2607.02770
synthetic dataset generation workflow using local file resources for finetuning llms.
Official implementation of MATPO: Multi-Agent Tool-Integrated Policy Optimization.
Building LLaMA 4 MoE from Scratch
transformers safetensors qwen3_next text-generation conversational license:apache-2.0
transformers safetensors qwen2 text-generation code qwen
Go BPE tokenizer (Encoder+Decoder) for GPT2 and GPT3
Implementation of a GPT-4o like Multimodal from Scratch using Python
The first pure SNN language model trained from scratch with a fully original architecture. 618M parameters • 93% sparsity • Runs on phone • Online learning via STDP • $260 total training cost
Perceive, Predict, Verify: Continual Pre-training for Multimodal Agentic Foundation Models
A RL approach to enable cost-effective, intelligent interactions between a local agent and a remote LLM
Continual harness optimization
transformers safetensors qwen text-generation custom_code zh
transformers safetensors qwen3_moe text-generation abliterated uncensored
transformers pytorch safetensors kclgpt text-generation codeshell
This repository contains a new adversarial training method for Generative Conversational Agents
A recipe that will walk you through using either Meta Llama 3.1 405B or OpenAI GPT-4o deployed on Azure AI to generate a synthetic dataset using UC Berkeley's Gorilla project RAFT method.
transformers gguf code codeqwen chat qwen
A VisualBasic(.NET) language kernel and runtime for scientific data computing, deep learning, LLM inference, GPU acceleration, visualization and command-line data-science applications — running on .NET (net10.0) across Windows, Linux and macOS.
🤖 language modeling sft demos
A framework for training world models with virtual environments, complete with annotated environment dataset (RetroAct), exploration agent (AutoExplore Agent), and GenieRedux-G - an implementation of Genie with enhancements
Training LLMs to reason and analyze data with notebooks
Agents, and RL environment, for optimizing GPU kernels on AMD ROCm using LLM agents. Benchmarks LLM serving workloads end-to-end, profiles bottleneck kernels, optimizes them via Claude Code or Codex, and scores on compilation, correctness, and speedup.
Compact and Agent-Native MoE Training System
transformers safetensors gguf gemma2 text-generation llama-factory
mlx safetensors gemma4_unified image-text-to-text gemma4 coding
Agentic Engineering for Reinforcement Learning
Training architecture for self-improving AI agents.
The official implementation of "ML-Agent: Reinforcing LLM Agents for Autonomous Machine Learning Engineering"
transformers pytorch safetensors llama text-generation japanese
Multi-agent synthetic data generation pipeline capable of generating and validating long horizon terminal/coding tasks for RL training
transformers safetensors llama text-generation text-generation-inference unsloth
transformers safetensors Emu3 text-generation custom_code arxiv:2409.18869
transformers safetensors qwen2 text-generation conversational en
Official implement on 'What and When to Distill: Selective Hindsight Distillation for Multi-Turn Agents'
transformers pytorch mpt text-generation custom_code dataset:glaiveai/glaive-function-calling
An opinionated atlas of post-R1 LLM post-training: RL, rewards, alignment, distillation, agents, and their connections.
Training an LLM to use a calculator with multi-turn reinforcement learning, achieving a **62% absolute increase in evaluation accuracy**.
transformers pytorch t5 text-generation arxiv:2109.00859 arxiv:2207.01780
transformers safetensors qwen2 text-generation sea multilingual
gguf code codeqwen chat qwen qwen-coder
transformers safetensors llama text-generation arxiv:2506.03524 license:mit
A PPO-style Agentic RL framework without a separate critic model and multiple rollouts
Implementation of Single-Agent and Multi-Agent Reinforcement Learning Algorithms. MATLAB.
StarCraft II Multi Agent Challenge : QMIX, COMA, LIIR, QTRAN, Central V, ROMA, RODE, DOP, Graph MIX
transformers safetensors qwen3_next text-generation abliterated uncensored
从对话数据到训练:数字分身 + 模型蒸馏 From Dialogue Data to Training Closed-Loop: Digital Twin + Model Distillation
[NeurIPS 2026] GenEnv: Difficulty-Aligned Co-Evolution Between LLM Agents and Environment Simulators
transformers safetensors internlm2 feature-extraction text-generation conversational
transformers safetensors llama text-generation conversational arxiv:2403.04652
transformers safetensors llava_cohere text-generation merge image-text-to-text
transformers safetensors qwen2 text-generation code qwen
Official code for the publication "Large Language Models as Zero-shot Dialogue State Tracker through Function Calling" https//arxiv.org/abs/2402.10466
A Universal Deep Reinforcement Learning Framework
This is a dataset intended to train a LLM model for a completely CVE focused input and output.
transformers pytorch llama text-generation facebook meta
transformers safetensors iquestcoder text-generation code industrial-code
transformers safetensors qwen3_moe text-generation conversational arxiv:2505.09388
Implementation of Multi-Agent Reinforcement Learning algorithm(s). Currently includes: MADDPG
A pluggable AutoML experiment agent for multi-task hyperparameter optimization
MUA-RL: MULTI-TURN USER-INTERACTING AGENT REINFORCEMENT LEARNING FOR AGENTIC TOOL USE
transformers pytorch gguf orion text-generation code
transformers safetensors llama text-generation gpt llm
Official implementation of WHALE (Weight-Harness Alternating Learning)
Watch Every Step! LLM Agent Learning via Iterative Step-level Process Refinement (EMNLP 2024 Main Conference)
An open-source implementaion for fine-tuning SmolVLM.
A multi-agent system trained with GRPO for reliable long-horizon task planning and execution.
gguf text-generation en dataset:PrimeIntellect/verifiable-coding-problems dataset:likaixin/TACO-verified dataset:livecodebench/code_generation_lite
transformers gguf gemma4 coding code reasoning
A unified framework for building agentic RL environments.
NeurIPS 2025: Structural Entropy Guided Agent for Detecting and Repairing Knowledge Deficiencies in LLMs