Upstar existe aussi en français.

Voir cette page en français

Upstar ranks the fastest-rising GitHub projects every hour, across 69 topics. Browse topics

#llm-serving · AI & ML

LLM serving

Inference engines, model servers and the gateways that route requests across model APIs.

This week (Oct 27 – Nov 2), the top-ranked are Strata (+4,674), antirion (+302) and resin (+205). The 151 ranked repos gained 27,414 stars, 49% fewer than the week before.

Updated Nov 2, 12:33 UTC · ranked every hour from GitHub’s daily star counts

Updated 51 min ago

GitHub topic ↗ (opens in new tab)

This week’s ranking

Showing 50 of 152

How Upstar ranks

Full ranking, with the repositories hidden by default. Back to the ranking

  1. 1
    Niko1221/Strata Qwen3.8-Flash-Next on any consumer hardware: one-click install for Windows / Linux. Strata inference engine, OpenAI/Anthropic API on localhost, optional image input. C++ C++ · 38.1k total
    Gained in 7 d +4.7k Score 25.6 Total 38.1k
  2. 2
    eliasbiondo/antirion 🕷️ The fastest open-source AI Gateway and LLM Gateway — one binary with built-in console to route, secure, budget and observe every LLM call across OpenAI, Anthropic, Gemini, Bedrock and more. Rust Rust · 622 total Spotted at 95 stars
    Gained in 7 d +302 Score 16.9 Total 622
  3. 3
    ekzhang/resin 🧱 Experimental library for building inference engines Python Python · 368 total Mixed signals
    Gained in 7 d +205 Score 16.1 Total 368
  4. 4
    StartLuxLabs/StartLux-Prism Fast, exact inference for StartLux-Decision models on NVIDIA GPUs: the released GGUF files on int8 and 4-bit tensor cores, behind the /v1/systemone API, with up to 5.3× lower latency than vLLM in FP8 and bit-identical answers. Python Python · 238 total
    Gained in 7 d +149 Score 14.9 Total 238
  5. 5
    jesdga95/basalt Local inference engine for Qwen3.8-Flash-Next on NVIDIA Blackwell C++ C++ · 442 total
    Gained in 7 d +214 Score 14.2 Total 442
  6. 6
    NerverDevs/nervecenter new-api / sub2api 中转站账号管理中心：余额、API Key、用量、倍率、模型比价 | https://nvcenter.cc TypeScript TypeScript · 198 total
    Gained in 7 d +137 Score 13.7 Total 198
  7. 7
    deepseek-ai/DeepInfer DeepInfer: High-performance, large-scale inference framework for DeepSeek models Python Python · 24.9k total
    Gained in 7 d +2.1k Score 13.7 Total 24.9k
  8. 8
    trymirai/uzu A high-performance inference engine for AI models Rust Rust · 3.2k total Healthy
    Gained in 7 d +597 Score 11.7 Total 3.2k
  9. 9
    Alice-Marx/dsh-model-router Model Router and GALGame plugin for DeepSeek Harness JavaScript JavaScript · 623 total Healthy
    Gained in 7 d +221 Score 11.0 Total 623
  10. 10
    skyzh/tiny-llm learn LLM inference system on Apple Silicon for systems engineers: build a tiny vLLM + Qwen Python Python · 5.5k total Healthy
    Gained in 7 d +730 Score 10.6 Total 5.5k
  11. 11
    VecSzn/PolyStrata Run large AI models on your own PC: Qwen3.8-Flash-Next on Strata's engine, and GLM 5.3 Flash, DeepSeek V4 Flash, LFM2.5, Qwen3.6, Ornith 1.5, Gemma 4, Tiel Coder and MiniCPM-V on a second engine written for PolyStrata. One-click install for Windows and Linux, OpenAI/Anthropic API on localhost. C++ C++ · 760 total Spotted at 196 stars
    Gained in 7 d +233 Score 10.1 Total 760
  12. 12
    mw00/project-maya Project Maya - run GLM-5.3-Flash (321B MoE), DeepSeek-V4.1-Flash-1017 (748B MoE) and MiniMax-M3.1-Flash (428B MoE) privately on your own GPU(s), from one card to 16, NVIDIA or AMD, Linux or Windows. Engine, server, dashboard, OpenAI- and Anthropic-compatible API, quant tools. Cuda Cuda · 2.4k total Spotted at 762 stars
    Gained in 7 d +438 Score 9.9 Total 2.4k
  13. 13
    respanai/respan-gateway Open-source LLM gateway in Rust. One endpoint for OpenAI, Anthropic, Gemini and every other vendor. Any SDK, any model, any tool. Rust Rust · 487 total
    Gained in 7 d +158 Score 8.7 Total 487
  14. 14
    cheaper-inference/OmniRoute Never stop coding. Free MIT AI gateway: one endpoint, 375 providers (150+ free), 1,400+ models Kimi, Claude, GPT, Gemini, GLM, DeepSeek, MiniMax. Works with Claude Code, Codex, Cursor, OpenCode, Cline & Copilot. Quota-aware auto-fallback, RTK+Caveman compression saves 15-95% tokens, MCP/A2A… TypeScript TypeScript · 83.7k total Healthy
    Gained in 7 d +2.4k Score 8.5 Total 83.7k
  15. 15
    drawthingsai/local-code-community Source code of Local Code, a private AI agent for Mac, iPad and iPhone. Runs DeepSeek 4.1 on device, with Python, Node.js and a shell built into the app. C C · 85 total
    Gained in 7 d +85 Score 8.5 Total 85
  16. 16
    FlashML-org/FreeToken FreeToken brings datacenter-scale model serving to your desktop. Run massive models locally, fast and efficiently. Python Python · 15.7k total Healthy
    Gained in 7 d +946 Score 7.8 Total 15.7k
  17. 17
    kekzl/imp Run large AI chat models privately on your own NVIDIA RTX 50 or RTX PRO graphics card. On one RTX 5090, Qwen3.8-27B writes about 70 words per second, 1.4x faster than vLLM. Even a model almost four times larger than the card's memory runs. Works with apps made for ChatGPT or Claude. Starts with one… C++ C++ · 124 total Healthy
    Gained in 7 d +77 Score 7.7 Total 124
  18. 18
    routerhubSI/uncensored-ai-api Uncensored AI API and models — one key for Claude, GPT, Gemini, Grok, DeepSeek, GLM and Qwen, with uncensored and abliterated variants where available. OpenAI- and Anthropic-compatible. Pay per token. 504 total
    Gained in 7 d +139 Score 7.3 Total 504
  19. 19
    Calcium-Ion/AstrLink A local AI gateway for AI agents, unifying your subscriptions and API providers with smart routing and on-device privacy protection. Go Go · 524 total
    Gained in 7 d +142 Score 7.3 Total 524
  20. 20
    synchopate/nemotron-30b-cpu-256k Running a 30B MoE LLM on CPU at 30-50+ tok/s with a 256K Context Window via Hybrid Mamba-2 SSM, L3 NTA Slicing, AVX-512 VNNI Tile Repacking & 2MB HugePages Python Python · 70 total New
    Gained in 7 d +70 Score 7.0 Total 70
  21. 21
    giienew/9router-pro 9Router Pro by Gienetic — Pro edition of 9Router with Prompt Injection (system prompt or chat script per global / API key / model), per-key token quotas, accent themes and an optimized dashboard. JavaScript JavaScript · 83 total
    Gained in 7 d +69 Score 6.9 Total 83
  22. 22
    ThinkWatchProject/ThinkWatch-Lite Local gateway for Claude Code, Codex and other AI clients on macOS, Windows and Linux: switch upstreams without touching clients, keep API keys from relays, cut off malicious tool calls, trace every request. TypeScript TypeScript · 532 total
    Gained in 7 d +134 Score 6.7 Total 532
  23. 23
    bitcloud-ai/Bittune Local-first engineering agent that deploys, benchmarks and tunes vLLM/SGLang inference services on your own GPUs, with auditable run evidence. TypeScript TypeScript · 161 total Mixed signals
    Gained in 7 d +66 Score 6.6 Total 161
  24. 24
    rednote-machine-learning/RedKnot Efficient Long-Context LLM Serving with Head-Aware KV Reuse and SegPagedAttention Python Python · 3.1k total Mixed signals
    Gained in 7 d +328 Score 6.3 Total 3.1k
  25. 25
    dayuwang401/spectral-orthogonal-exploration Official code for "Student Guides Teacher: Weak-to-Strong Inference via Spectral Orthogonal Exploration" (SOE, ACL 2026 Oral). A training-free, test-time scaling method that fixes reasoning collapse in LLM mathematical reasoning via orthogonal probing. Python Python · 134 total
    Gained in 7 d +60 Score 6.0 Total 134
  26. 26
    mnfst/awesome-free-llm-apis List of Permanent Free LLM API (API Keys) JavaScript JavaScript · 12k total Mixed signals
    Gained in 7 d +636 Score 6.0 Total 12k
  27. 27
    cablate/model-relay Per-window model switcher, subagent router, and automatic 429 failover for Claude Code — switch any window with /provider, delegate subagents across models, and keep Remote Control & statuslines working. JavaScript JavaScript · 103 total
    Gained in 7 d +58 Score 5.8 Total 103
  28. 28
    Wallawalla47/Infernix C++/CUDA single-GPU inference engine for Qwen3.8 (incl. Qwen3.8-Flash-Next with offloaded experts) on the RTX 5090; grew from NInfer C++ C++ · 218 total
    Gained in 7 d +70 Score 5.8 Total 218
  29. 29
    martina9516/uncensored-ai-models Uncensored AI models from RouterHub — the website for one-key API access to Claude, GPT, Gemini, Grok, DeepSeek, GLM and Qwen, with uncensored and abliterated variants where available. OpenAI- and Anthropic-compatible. Pay per token. 108 total New
    Gained in 7 d +55 Score 5.5 Total 108
  30. 30
    01men/ybkkasi One control plane for every AI asset — MCP servers, skills, agents, AI apps. Self-hosted ledger with lifecycle state machine, 85 permission points, department × agent × skill cost attribution, and audit trail. MIT, zero runtime deps (Node 22 + node:sqlite). 企业 AI 资产治理。 JavaScript JavaScript · 58 total
    Gained in 7 d +53 Score 5.3 Total 58
  31. 31
    musicurgy/delta-inference-engine Delta Inference Engine: DeepSeek-V4.1-Flash on one NVIDIA GPU, natively on Windows (OpenAI-compatible server) C++ C++ · 53 total New
    Gained in 7 d +53 Score 5.3 Total 53
  32. 32
    shuishuipingan/qoder2api-hub Qoder国际国内多账号反代网关，Qoder 国际国内双区多账号 OpenAI 兼容网关中枢 — COSY 签名推理、稳定设备指纹防风控、OAuth 设备授权免客户端登录、每日签到与 Pro 福利、后台定时调度器、Web 监控看板，支持 Codex / Claude Code 与标准 OpenAI 客户端 Python Python · 498 total
    Gained in 7 d +105 Score 5.3 Total 498
  33. 33
    nace-ai/kapitsa Blazing fast Apple Silicon optimized inference engine built for Qwen & Nemotron. OpenAI + Anthropic API compatible. Built by folks from nace.ai C++ C++ · 595 total
    Gained in 7 d +114 Score 5.2 Total 595
  34. 34
    syv-ai/HyperQwen Serve large Qwen models fast on the GPUs you actually own. Qwen3.8-27B on a single 24 GB card with vLLM: 127 tok/s single-user (381 when the answer quotes the prompt), ~1,035 tok/s at 64 concurrent, 150k-262k context. vLLM patches, requant pipeline, benchmarks. Python Python · 2.5k total Healthy
    Gained in 7 d +245 Score 5.2 Total 2.5k
  35. 35
    Generated-Computers/Locals Run many AI agents at once on your Mac, up to 3x faster. Python Python · 447 total
    Gained in 7 d +94 Score 5.0 Total 447
  36. 36
    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. C C · 574 total
    Gained in 7 d +106 Score 4.9 Total 574
  37. 37
    ahtx/kindi A custom LLM inference engine for low-cost GPUs: NVIDIA cards from a 1 GB Quadro K420 to an RTX 3080, the AMD Radeon 780M and the CPU. Runs the Bonsai 2B to 27B models and Nemotron with an OpenAI-compatible server. Python Python · 113 total
    Gained in 7 d +48 Score 4.8 Total 113
  38. 38
    ddalcu/mlx-serve Native LLM inference server for Apple Silicon. OpenAI + Anthropic API compatible. No Python. Zig backend, Swift frontend macOS app with chat, music, voice, video generation. Zig Zig · 2.3k total Healthy
    Gained in 7 d +217 Score 4.7 Total 2.3k
  39. 39
    T-Crypt/berth Berth is a lightweight GPU resource manager written in Go. It controls which applications get access to your GPU, when they get it, and how that access is handed off. Go Go · 279 total
    Gained in 7 d +67 Score 4.6 Total 279
  40. 40
    carloslfu/slotstream Run a 105 GB AI model on a Mac that can't hold it. Qwen3.8-Flash-Next 4-bit, streamed from SSD on 16–64 GB Apple Silicon Macs. Fully native Swift, MLX and Metal. Offline, no Python. CLI, Swift library, and Ollama/OpenAI/Anthropic APIs. Works with Claude Code and Codex. Swift Swift · 642 total Healthy
    Gained in 7 d +104 Score 4.5 Total 642
  41. 41
    Indras-Mirror/Xolotl Xolotl (formerly Strata-DSV4.1): run DeepSeek-V4.1-Flash, all 552B, as a local coding agent on one gaming PC (RTX 4090 + 90 GB RAM) at up to ~27 tok/s, 512K context, October 1017 builds included. Plus GLM-5.3-Flash and Qwen3.8. MoE experts tiered across VRAM, pinned RAM and NVMe; OpenAI + Anthropic… C++ C++ · 441 total
    Gained in 7 d +83 Score 4.4 Total 441
  42. 42
    incoai/splash A local inference engine for Apple silicon, built around the model. C++ C++ · 2k total
    Gained in 7 d +184 Score 4.3 Total 2k
  43. 43
    lyogavin/airllm AirLLM 2.8T LLM inference with single 4GB GPU Jupyter Notebook Jupyter Notebook · 41.6k total Healthy
    Gained in 7 d +831 Score 4.1 Total 41.6k
  44. 44
    MemTensor/MemT-1.0 Official repository for MemT-1.0 Python Python · 498 total
    Gained in 7 d +83 Score 4.1 Total 498
  45. 45
    magnitudedev/magnitude Open source inference engine that tunes its kernels on your hardware to find the fastest way to run each model. 2x faster than Ollama. Works on Apple Silicon, NVIDIA, AMD, Intel, or just a CPU. Rust Rust · 7.7k total Healthy
    Gained in 7 d +334 Score 3.9 Total 7.7k
  46. 46
    chalk-ai/chalk-router-oss chalk-router: an extremely fast & focused LLM router implementation. Rust Rust · 39 total New
    Gained in 7 d +39 Score 3.9 Total 39
  47. 47
    eagle-hq/EAGLE-4 Official Inference Code of EAGLE-4 Python Python · 1.1k total
    Gained in 7 d +113 Score 3.6 Total 1.1k
  48. 48
    WizisCool/oh-my-cpa Self-hosted control plane & usage observability console for CLIProxyAPI (CPA) — AI providers, client keys, streaming telemetry, and model cost snapshots. TypeScript TypeScript · 296 total
    Gained in 7 d +51 Score 3.3 Total 296
  49. 49
    3145544645/gouji 构屎 GOUJI — 纯 C++/CUDA 的三值 mini LLM。2 bit/权重，零 Python；联想记忆工具路由，零训练参数；143 个缺陷与 29 个变异测试记录在案。 C++ C++ · 99 total
    Gained in 7 d +32 Score 3.2 Total 99
  50. 50
    christopherowen/spark-ds41f Reproducible Docker/vLLM deployment, tuning, and benchmarks for DeepSeek V4.1 Flash on a switchless three-node or four-node DGX Spark fabric. Python Python · 78 total
    Gained in 7 d +31 Score 3.1 Total 78
Show more

Emerging in LLM serving Past weeks: Oct 26Oct 19