qwen3 repositories
Trending repositories tagged qwen3, ranked by trending score. Updated daily.
-
1
vllm-project/vllm
A high-throughput and memory-efficient inference and serving engine for LLMs
GitHub · Python · 97,308 stars · 25,102 forks · +97 stars in 24 hours · Trending score 4.20
-
2
microsoft/AesCode
Aesthetic code generation with decoupled cross-modal rewards: turning prompts into editable, verifiable HTML/CSS infographics, with image generators as aesthetic guidance.
GitHub · Python · 284 stars · 10 forks · +201 stars in 24 hours · Trending score 3.92
-
3
adventists-ai/duplexjev
DuplexJev-32B and DuplexJev-4B: speech decision models for full-duplex voice agents. Typed single-token readout from a frozen speech LLM, served with vLLM. Try it: api.adventists.cn/duplexjev
GitHub · Python · 156 stars · 4 forks · +124 stars in 24 hours · Trending score 3.68
-
4
synchopate/llama-modchip
Zero-GPU CPU inference for 30B-284B MoE models (Qwen3.8 176.5B cold-cache certified in 28.26/48.35 GiB Cgroup v2, DeepSeek-V4 284B, Nemotron-30B) + RouteCrack 2.0.
GitHub · Python · 1,398 stars · 137 forks · +33 stars in 24 hours · Trending score 3.35
-
5
HoppouAI/Breeze-TTS-2.cpp
Bilingual English and Mandarin text to speech in C++ and GGUF. Voice design, cloning, direction and experimental voice conversion. Vulkan backend, streaming HTTP and WebSocket server, web UI and a C API.
GitHub · C++ · 196 stars · 34 forks · +67 stars in 24 hours · Trending score 3.19
-
6
jaredpalmer/kev
Jev-like family of decision models built on top of Qwen3.5/3.8 you can train and run on your own
GitHub · Python · 10,078 stars · 690 forks · +26 stars in 24 hours · Trending score 2.81
-
7
sangharshadhyeta/Strata-HotRod
Run Qwen3.8-Flash-Next (125B) on an RTX 3060 12 GB with 30 GB of RAM, streaming experts from the SSD. 50-70 tok/s, 262K context. A fork of Strata.
GitHub · C++ · 68 stars · 9 forks · +25 stars in 24 hours · Trending score 2.72
-
8
peonist-ai/halogen-flash-server
Local inference that behaves like a frontier API: Qwen3.8-Flash-Next on AMD Strix Halo plus small NPU models, over the OpenAI and Anthropic APIs on one port.
GitHub · Shell · 1,412 stars · 103 forks · +17 stars in 24 hours · Trending score 2.61
-
9
AEON-7/Qwen3.6-27B-AEON-Ultimate-Uncensored-DFlash
Fully uncensored, capability-enhanced abliteration of Qwen3.6-27B. NVFP4 + z-lab DFlash speculative decoding (n=12) on the unified ghcr.io/aeon-7/aeon-vllm-ultimate:latest container, tuned for long-context draft acceptance on DGX Spark. 6 HF variants (BF16/NVFP4/MTP/MTP-XS), docker-compose, and QuickStart.
GitHub · Python · 500 stars · 57 forks · +27 stars in 24 hours · Trending score 2.44
-
10
modelscope/ms-swift
Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.8, DeepSeek-V4, GLM-5.1, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM5.3, Gemma4, Llava, Phi4, ...) (AAAI 2025).
GitHub · Python · 16,137 stars · 1,820 forks · +11 stars in 24 hours · Trending score 2.41
-
11
JakeATX/llamAmpere
llama.cpp fork for significantly improved performance on Ampere generation Nvidia cards (especially RTX 3090 / 3090 Ti): example: 110+ tok/s over a 100K-token generation at temperature 1 for qwen3.8, supporting full 262K ctx. Supports low memory (12gb) configs for 200k+ ctx. 120+ TPS, 3300+ prefill for QFN 3.8.
GitHub · C++ · 405 stars · 50 forks · +8 stars in 24 hours · Trending score 2.34
-
12
1Panel-dev/MaxKB
🔥 MaxKB is an open-source platform for building enterprise-grade agents. 强大易用的开源企业级智能体平台。
GitHub · Python · 23,183 stars · 3,229 forks · +13 stars in 24 hours · Trending score 2.29
-
13
qualcomm/GenieX
Run frontier LLMs and VLMs locally on Qualcomm devices across NPU, GPU, and CPU with a few lines of code
GitHub · Go · 8,541 stars · 1,091 forks · +7 stars in 24 hours · Trending score 2.05
-
14
NVIDIA-NeMo/Automodel
🚀 Pytorch Distributed native training library for LLMs/VLMs with OOTB Hugging Face support
GitHub · Python · 1,097 stars · 390 forks · +9 stars in 24 hours · Trending score 2.03
-
15
yoloshii/gigaxity-deep-research
Open-source deep research MCP. Qwen3-30B-A3B-Thinking via OpenRouter, cited web synthesis for Claude Code, Codex, Cursor, Hermes and any MCP-compatible agent.
GitHub · Python · 66 stars · 9 forks · +14 stars in 24 hours · Trending score 1.98
-
16
hec-ovi/vllm-awq4-qwen
vLLM Qwen 3.6-27B (AWQ-INT4) + DFlash speculative decoding on AMD Strix Halo (gfx1151 iGPU, 128 GB UMA, ROCm 7.13). 24.8 t/s single-stream, vision, tool calling, 256K context, OpenAI-compatible, Docker. Matches DGX Spark FP8+DFlash+MTP at a third of the cost. No CUDA.
GitHub · Python · 53 stars · 5 forks · +8 stars in 24 hours · Trending score 1.60
-
17
ShuaixinHuang/image-multiple-angles-3d-camera
🎬 Generate images from any camera viewpoint via 3D interactive control. Drag the camera in 3D space or use sliders to set azimuth/elevation/distance, then generate. Built with Three.js + Gradio, bilingual ZH/EN UI.
GitHub · Python · 1,005 stars · 21 forks · 0 stars in 24 hours · Trending score 1.50
-
18
lawrence3699/jev-style
JevStyle: local decision models for AI agents. Classify, route and score with a probability per option. systemone-compatible local server, MCP tools, Claude Code guard. Open weights, Apache-2.0.
GitHub · Python · 138 stars · 4 forks · +1 stars in 24 hours · Trending score 1.47
-
19
dwim-sh/dwim
dwim is a coding agent that you control. Local by default.
GitHub · Rust · 44 stars · 2 forks · +6 stars in 24 hours · Trending score 1.45
-
20
shyringo/qwen3.8-flash-next-in-c
The best model under 200B at near 10 token/s exact batch throughput on a single laptop CPU, with an automatic 8 GB RAM path. Native C, no GPU or Python. | 200B 以下最强模型,单颗笔记本 CPU 精确批处理接近 10 token/s,最低 8 GB 内存自动运行。原生 C 语言,无需 GPU 或 Python。
GitHub · C · 24 stars · 8 forks · +6 stars in 24 hours · Trending score 1.45
-
21
stdlogits/neuron-lens
A catalog of every MLP neuron and attention head of Qwen3-8B and Gemma 4 12B, read from the weights through two Jacobian lenses (R and S), from activations, and from their exact token drivers
GitHub · Python · 60 stars · 5 forks · +5 stars in 24 hours · Trending score 1.44
-
22
Quantatirsk/AsrServe
Self-hosted realtime and offline ASR with SOTA open models, speaker diarization and an OpenAI-compatible API. GPU, CPU, macOS Apple Silicon.
GitHub · Python · 362 stars · 56 forks · +7 stars in 24 hours · Trending score 1.35
-
23
prabanta-dev/janas
37 tok/s from Qwen3-Next-80B (48 GB) on a 32 GB laptop, 24 without drafts vs 2.8 for llama.cpp; reading 62 vs 16 tok/s. Faster than llama.cpp on all 13 models compared in Q4_K_M. LLMs bigger than your RAM, pure C, CPU or GPU.
GitHub · C · 735 stars · 73 forks · 0 stars in 24 hours · Trending score 1.28
-
24
pegainfer-project/pegainfer
Pure Rust + CUDA LLM inference engine — no PyTorch, OpenAI-compatible, serves Qwen3 to Kimi-K2
GitHub · Rust · 748 stars · 122 forks · +3 stars in 24 hours · Trending score 1.23