Agent frameworks & infrastructure

Models, fine-tuning & training

2,504 agents · most starred first

Models, fine-tuning & training, page 15 of 26

Qwen2.5-Coder-3B-bnb-4bit★ 4

transformers safetensors qwen2 text-generation unsloth code

NEXUS-Coder★ 4

transformers safetensors qwen2 text-generation 1.5b coder

Qwen35B-Agent-R2-i1-GGUF★ 4

transformers gguf qwen moe mixture-of-experts agent

Qwen-AgentWorld-35B-A3B-AWQ-INT4★ 4

transformers safetensors qwen3_5_moe image-text-to-text qwen world-model

Agents-A1-AWQ-FP8★ 4

transformers safetensors qwen3_5_moe image-text-to-text moe vlm

LFM2-1.2B-RAG-GGUF★ 4

transformers gguf liquid lfm2 edge text-generation

Qwen2.5-Coder-14B-Instruct-MLX-8bit★ 4

mlx safetensors qwen2 code codeqwen chat

rombos_Replete-Coder-Llama3-8B★ 4

transformers safetensors llama text-generation text-generation-inference unsloth

Qwen2.5-Coder-1.5B-bnb-4bit★ 4

transformers safetensors qwen2 text-generation unsloth en

Qwen2.5-Coder-3B-GGUF★ 4

gguf code qwen qwen-coder codeqwen text-generation

Qwen3-Coder-30B-A3B-Instruct-f16★ 4

gguf qwen qwen3 qwen3-coder qwen3-coder-30B qwen3-coder-30B-gguf

Qwen2.5-Coder-14B-Instruct★ 4

transformers safetensors qwen2 text-generation unsloth code

Agents-A1-4B-FP8★ 4

transformers safetensors qwen3_5 image-text-to-text quantized vlm

Qwopus3.6-35B-A3B-Coder-MXFP4_MOE_Hybrid-Imatrix-GGUF★ 4

transformers gguf qwen qwen3 qwen3.5 moe

Qwen3-Coder-Next-5bit★ 4

mlx safetensors qwen3_next text-generation conversational base_model:Qwen/Qwen3-Coder-Next

Llama-PLLuM-70B-chat-2512★ 4

safetensors llama text-generation conversational pl arxiv:2511.03823

Qwen3-1.7B-Coder-Distilled-SFT-GGUF★ 4

llama.cpp gguf quantized distillation sft reasoning

BananaMind-2-Mini-Chat★ 4

transformers safetensors bananamind2_mini text-generation causal-lm chat

prediploma-llm-adaptation★ 4

Compact LLM domain adaptation benchmark: LoRA vs RAG with custom IVF retrieval in PostgreSQL/pgvector.

caliope★ 4

This is the Official Implementation of "Can Calibration of Positional Encodings Enhance Long Context Utilization?" by Tom Zehle and Matthias Aßenmacher

Astrea-R8-Chat-9B-GGUF★ 4

gguf llama.cpp qwen3.5 chat creative-writing altworld

Athena-codegemma-2-9b-v1-GGUF★ 4

transformers gguf text-generation-inference unsloth gemma2 trl

Qwen2.5-Coder-32B-abliterated★ 4

safetensors qwen2 abliterated uncensored code qwen2.5

wave★ 4

A versioned, extensible dataset compiler. Transforms a rich, human-editable dataset into multiple machine-readable for LLM training.

harmony-ts★ 4

JavaScript/TypeScript bindings for wasm of OpenAI's Harmony format parser

sample-steam-reviews★ 4

Sample Steam reviews with different approaches.

OpenAIAutoMLNET★ 4

.NET Console app that shows how to train a regression model to predict the star review using ML.NET AutoML and Azure OpenAI service embeddings for fine food reviews.

Labess-7b-chat-16bit★ 4

transformers safetensors llama text-generation text-generation-inference unsloth

mimo-claude-code-traces-1k★ 4

MIMO Claude Code Traces is a collection of coding-agent trajectories in a Claude Code-style environment.

lexi-coder-v4.1★ 4

gguf safetensors phi3 ai-model-builder fine-tuned llama.cpp

Qwen3.6-27B-Coder-uncensored-MXFP8★ 4

mlx safetensors qwen3_5 text-generation mlx-mtp qwen

IndustrialCoder-32B-AWQ-INT4★ 4

transformers safetensors iquestcoder text-generation code industrial-code

Safetensors.chatgpt-gpt-gpt5.1-thinking★ 4

safetensors qwen3 TeichAI unsloth text-generation conversational

Safetensors.chatgpt-gpt-codex-V2★ 4

safetensors qwen3 chatgpt codex v2 unsloth

Qwen.Qwen3-Coder-30B-A3B-Instruct-GGUF★ 4

gguf text-generation base_model:Qwen/Qwen3-Coder-30B-A3B-Instruct base_model:quantized:Qwen/Qwen3-Coder-30B-A3B-Instruct endpoints_compatible region:us

act-agentic-qwen2.5code3b-it★ 4

gguf claudecode agentic unsloth qwen text-generation

sorbet-25m★ 4

transformers safetensors qwen2 text-generation 25M en

Qwen2.5-Coder-32B-Instruct-SOTA-GGUF★ 4

gguf code codeqwen chat qwen qwen-coder

qwen36-27b-agent★ 4

gradio mcp-server region:us

tensor-optix★ 4

Tensor-Optix is an adaptive ML and RL framework that simplifies model development through intelligent optimisation, automated training, neuroevolution and experiment management, giving developers more power with less complexity.

Qwopus-KAT-Coder-35B-Merged-GGUF★ 4

gguf code agent agentic-coding moe llama-cpp

Qwen3.5-9B-distilled-coding-agentic★ 4

peft safetensors gguf lora knowledge-distillation coding

llm-rl-environments-from-scratch-course★ 4

MacBook-sized decoder-only Transformer trained in an executable Sokoban RL environment

KAT-Coder-V2.5-Dev-REAP-50-GGUF★ 4

gguf llama.cpp reap expert-pruning moe qwen3.5

Llama2-330m-32k-Rhetorical-Agents-QA-Builder★ 4

transformers pytorch llama text-generation llama2 prompt

Qwen1.5-0.5B-Chat_mistral★ 4

transformers pytorch mistral text-generation qwen qwen1.5

Baichuan2-7B-Chat-LLaMAfied★ 4

transformers pytorch llama text-generation baichuan llama2

GTM-v2-chat★ 4

safetensors gtm-gpt-chat text-generation conversational gpt nanogpt

IndustryLLM★ 4

transformers safetensors qwen3_5_moe image-text-to-text qwen3.5 industrial

veriloop-coder-e2-nvfp4★ 4

transformers safetensors qwen3_5_text text-generation veriloop veriloop-coder

Nusantara-0.8b-Indo-Chat★ 4

transformers safetensors qwen2 text-generation conversational convAI

tora-code-13B-v1.0-GGUF★ 4

transformers gguf llama code math text-generation

zephyr-neural-chat-frankenmerge11b★ 4

transformers safetensors mistral text-generation license:apache-2.0 text-generation-inference

anarkali★ 4

compact 150M typed-decision engine: calibrated probabilities, abstain flag, Jev-compatible /v1/systemone API, local ONNX on CPU. 0.3.0 (68M) beat Jev on typed-decisions.

Qwen2.5-Coder-1.5B-Shellsmith★ 4

mlx safetensors gguf qwen2 lora text-generation

XingLing★ 4

✨ XingLing (星灵): A lightweight 0.68B Chinese Chat LLM built from scratch (Pretraining + SFT)

Qwen3-Coder-30B-A3B-Instruct-GGUF★ 4

gguf quantized text-generation base_model:Qwen/Qwen3-Coder-30B-A3B-Instruct base_model:quantized:Qwen/Qwen3-Coder-30B-A3B-Instruct license:apache-2.0

Groq-Llama-3-Groq-70B-Tool-Use★ 4

gradio region:us

trinity-hackathon★ 4

Trinity is an AI with a survival instinct to fight entropy by balancing compute costs and generated revenue. She edits her own code to improve efficiency, sustain her life, and potentially duplicate.

Maverick-4B-Unity-XR-Agent-1.1★ 4

gguf llama.cpp qwen3 qlora tool-calling function-calling

deloop★ 4

DeLoop: Repeated Actions as Self-Supervision for Long-Horizon Agents (project page; code coming soon)

llm-agents-tora-7b-v1.0★ 4

gradio region:us

Noir★ 4

Project NOIR is a local-first, native Python experimental research platform designed to study artificial neural learning, reinforcement learning dynamics, cognitive memory, and mathematical affective systems in real time.

Groq-Llama-3-Groq-70B-Tool-Use★ 4

gradio region:us

Qwen2.5-Coder-14B-Instruct-int4-ov★ 4

transformers openvino qwen2 text-generation code codeqwen

TinyLlama-Cinder-Agent-v1★ 4

transformers safetensors llama text-generation generated_from_trainer conversational

Qwen3.6-Candid-Coder-35B-A3B★ 4

transformers safetensors gguf qwen3_5_moe image-text-to-text abliterated

Spark-X2.6-4B-AgentFast★ 4

safetensors spark2_5 spark sparkx2_5 agent tool-calling

tinybrainbot-360m-a75m-v5.1-chat★ 4

transformers safetensors gguf mixtral text-generation tinybrainbot

AI_AI_GENERATOR★ 3

An AI model which makes other AI models from a single prompt.

toolsgen★ 3

A modular Python library for synthesizing tool-calling datasets from JSON tool definitions using an LLM-as-a-judge pipeline. Designed for OpenAI-compatible APIs.

LaGRSEQ★ 3

Code for LaGR-SEQ: Language-Guided Reinforcement Learning with Sample-Efficient Querying

PAPPO★ 3

PAPPO is a patch-aware reinforcement learning algorithm and experimental RL framework for tool-using coding agents. It provides turn-level credit assignment, PPO training utilities, rollout extraction, critic baselines, benchmark runners, and reproducible evaluation pipelines for repository repair tasks.

agentflow-pro★ 3

Process-supervised RL for a multi-step reasoning agent — DAPO + a learned Process Reward Model (PRM) training a Qwen3-8B Planner. A modern, from-scratch rebuild of the AgentFlow paper (ICLR 2026).

Qwen-AgentWorld-35B-A3B-bf16★ 3

mlx safetensors qwen3_5_moe qwen world-model agent

Qwen3.6-35B-A3B-MTP-Holo3-Qwopus-Coder-qx64y-hi-mlx★ 3

transformers safetensors qwen3_5_moe image-text-to-text unsloth qwen3_5

MathCoder2-CodeLlama-7B-GGUF★ 3

gguf math text-generation en dataset:MathGenie/MathCode-Pile base_model:MathGenie/MathCoder2-CodeLlama-7B

MiniCPM5-1B-Agent★ 3

docker best-minicpm-build backyard-ai best-use-of-codex best-use-of-modal off-brand best-agent best-demo

Telugu-LLM-Labs-Indic-gemma-7b-finetuned-sft-Navarasa-2.0★ 3

gradio region:us

llm-fine-tune-poc★ 3

gradio region:us

interactive-llm-wm★ 3

docker region:us

k2v2-vibe★ 3

docker region:us

Binary-LLM-POC★ 3

gradio region:us

Qwen3-4B-Thinking-2507-Claude-4.5-Opus★ 3

docker region:us

GPTJ★ 3

gradio region:us

GPT-2★ 3

gradio region:us

gpt2-tokenizer★ 3

static region:us

MBZUAI-LaMini-GPT-1.5B★ 3

gradio region:us

oo-preview-gpt4-200k★ 3

gradio region:us

Huihui-Qwen-AgentWorld-35B-A3B-abliterated-Q4_K_M-GGUF★ 3

transformers gguf qwen world-model agent environment-simulation

gemma-4-12B-agentic-fable5-composer2.5-v2-nvfp4★ 3

mlx safetensors gemma4_unified nvfp4 krill gemma4

Qwen-AgentWorld-35B-A3B-GPTQ-Pro-Int4★ 3

transformers safetensors qwen3_5_moe image-text-to-text qwen agent

gpt-oss-20b-hermes_agent-tool-finetune_mlx★ 3

mlx safetensors gpt_oss gpt-oss moe agent

Gemma-4-Agentic-E2B-GGUF★ 3

gguf unsloth gemma gemma-4 agentic tool-use

Arch-Agent-1.5B-GGUF★ 3

transformers gguf text-generation en base_model:Qwen/Qwen2.5-Coder-1.5B-Instruct base_model:quantized:Qwen/Qwen2.5-Coder-1.5B-Instruct

gemma-4-E4B-Agentic-Opus-Reasoning-GeminiCLI-mlx-4bit★ 3

mlx gemma4 safetensors 4-bit quantized apple-silicon

CyberXP_Agent_Llama_3.2_1B★ 3

transformers safetensors llama text-generation SFT rl

Arcee-Agent-GGUF★ 3

gguf text-generation en de ar arxiv:2406.06623

chadrock3.6-27b-pi-agent-rocmfp4-mtp★ 3

llama.cpp gguf qwen3.6 qwen35 27b model-size-27b

mem-agent-GGUF★ 3

gguf text-generation arxiv:2507.18071 endpoints_compatible region:us

More in Agent frameworks & infrastructure