# Language models: what’s taking off on GitHub

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score. Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.

Data as of 2026-10-29 13:58 UTC, updated every 30 minutes.

1. **togg53192-cmd/jailbreaks** — 8.1 · Breakout · Language models · +1,029 stars measured 2026-10-29, 00:02–13:42 UTC, ≈ 1,732 by evening
   A collection of 31 system prompts, persona locks, and AGENTS.md files for 25+ models (Claude, Gemini, Grok, GLM, DeepSeek, Kimi, Qwen, Gemma, GPT-OSS). Intended for research on how instruction hierarchies and policy overrides affect model…
   Full card: https://gitnova.dev/en/r/togg53192-cmd/jailbreaks.md
2. **Edge0-AI/Edge0** — 7.3 · Breakout · Language models · Python · +1,231 stars measured 2026-10-29, 00:05–13:42 UTC, ≈ 1,864 by evening
   An open-source streaming MoE inference framework: expert weights are offloaded from SSD on demand while a trained prerouter predicts routing ahead of time. Runs on Apple Silicon via MLX and ships with two ready-to-run model tiers (35B and…
   Full card: https://gitnova.dev/en/r/Edge0-AI/Edge0.md
3. **OpenBMB/MiniCPM-OCR** — 6.8 · Early signal · Language models · Python · +213 stars measured 2026-10-29, 00:03–13:42 UTC, ≈ 355 by evening
   MiniCPM-OCR is an open-source OCR model family for parsing Chinese and English documents and PDFs into Markdown or JSON, with a 2B model (Jev) that scores recognition quality per region.
   Full card: https://gitnova.dev/en/r/OpenBMB/MiniCPM-OCR.md
4. **thomasunise/casecite** — 5.9 · Early signal · Language models · Python · +53 stars measured 2026-10-29, 00:03–13:43 UTC, ≈ 89 by evening
   Self-hosted legal research platform: upload your firm's documents, ask questions with citations, review contracts, draft documents and research judges. Runs on your own infrastructure with your own API keys.
   Full card: https://gitnova.dev/en/r/thomasunise/casecite.md
5. **NVIDIA/auto-ontology** — 5.4 · Early signal · Language models · Python · +57 stars measured 2026-10-29, 00:03–13:43 UTC, ≈ 106 by evening
   NVIDIA's Python framework that builds a semantic ontology graph over enterprise metadata, giving LLM agents trusted metric and entity definitions for reliable text-to-SQL.
   Full card: https://gitnova.dev/en/r/NVIDIA/auto-ontology.md
6. **lyramakesmusic/neuron-explorer** — 5.4 · Peaking · Language models · Python · +61 stars measured 2026-10-29, 00:03–13:43 UTC, ≈ 110 by evening
   A steering-first neuron explorer for Qwen3.5-4B: it pushes every MLP neuron up and down on prompts, keeps all rollouts, and lays the results out over a 3D view of the model for interpretability research.
   Full card: https://gitnova.dev/en/r/lyramakesmusic/neuron-explorer.md
7. **google-deepmind/detectorscope** — 5.2 · Early signal · Language models · Python · +37 stars measured 2026-10-29, 00:03–13:43 UTC, ≈ 71 by evening
   A toolkit for building AI-text detectors whose decisions are mechanistically grounded and auditable: each prediction is a sum of named features localized in the text. Built on sparse autoencoders and a concept bottleneck over the language…
   Full card: https://gitnova.dev/en/r/google-deepmind/detectorscope.md
8. **SoMarkAI/OXR** — 4.9 · Breakout · Language models · Python · +212 stars measured 2026-10-29, 00:02–13:43 UTC, ≈ 355 by evening
   OXR is an open-source document parsing model that turns PDFs and document images into Markdown or JSON with layout coordinates, recognizing 13 element types such as tables, formulas, chemical structures and code. With 1.3B parameters it…
   Full card: https://gitnova.dev/en/r/SoMarkAI/OXR.md
9. **baidu/Unlimited-OCR** — 4.8 · Breakout · Language models · Python · +171 stars measured 2026-10-29, 00:03–13:43 UTC, ≈ 297 by evening
   Baidu's OCR model for document parsing: extracts text from single images, multi-page files and PDFs in one pass, supporting long context up to 32768 tokens.
   Full card: https://gitnova.dev/en/r/baidu/Unlimited-OCR.md
10. **Niko1221/Strata** — 4.6 · Steady · Language models · C++ · +459 stars measured 2026-10-29, 00:03–13:43 UTC, ≈ 782 by evening
   A C++ local inference engine that runs the 125B MoE model Qwen3.8-Flash-Next on a regular PC with one NVIDIA GPU (12-24 GB) and 64 GB RAM, exposing an OpenAI/Anthropic-compatible API on localhost.
   Full card: https://gitnova.dev/en/r/Niko1221/Strata.md
11. **Tencent/WeVisOne-VL** — 4.5 · Early signal · Language models · Jupyter Notebook · +43 stars measured 2026-10-29, 00:04–13:43 UTC, ≈ 72 by evening
   WeVisOne-VL is a vision-language model by the WeChat Vision Team at Tencent that unifies object-level and pixel-level perception: detection, referring localization, counting, pointing, OCR, segmentation, and alpha matting. It handles…
   Full card: https://gitnova.dev/en/r/Tencent/WeVisOne-VL.md
12. **JustVugg/colibri** — 4.3 · Breakout · Language models · C · +171 stars measured 2026-10-29, 00:03–13:43 UTC, ≈ 305 by evening
   A pure-C, zero-dependency inference engine for running large MoE models (744B–2.8T parameters) on consumer hardware by treating VRAM, RAM, and storage as a single multitier hierarchy and streaming experts from disk.
   Full card: https://gitnova.dev/en/r/JustVugg/colibri.md
13. **deepseek-ai/DeepInfer** — 4.0 · Cooling · Language models · Python · +249 stars measured 2026-10-29, 00:04–13:44 UTC, ≈ 448 by evening
   Offline inference framework for running DeepSeek models at large scale: optimized kernels, MoE/Attention parallelism, deterministic sampling and structured outputs. Aimed at teams deploying DeepSeek-V4.1-Flash on NVIDIA SM100 or Huawei…
   Full card: https://gitnova.dev/en/r/deepseek-ai/DeepInfer.md
14. **QuantumNous/new-api** — 3.6 · Steady · Language models · Go · +69 stars measured 2026-10-29, 00:05–13:45 UTC, ≈ 114 by evening
   Self-hosted AI gateway that aggregates multiple LLM providers behind a unified OpenAI-, Claude-, or Gemini-compatible API with routing, access, and cost management.
   Full card: https://gitnova.dev/en/r/QuantumNous/new-api.md
15. **BerriAI/litellm** — 3.6 · Steady · Language models · Python · +53 stars measured 2026-10-29, 00:05–13:45 UTC, ≈ 90 by evening
   Open-source AI gateway and Python SDK that calls 100+ LLM providers (OpenAI, Anthropic, Bedrock, Azure, VertexAI, etc.) through a single OpenAI-compatible interface with cost tracking, guardrails, load balancing, and logging.
   Full card: https://gitnova.dev/en/r/BerriAI/litellm.md
16. **p-e-w/heretic** — 3.5 · Steady · Language models · Python · +61 stars measured 2026-10-29, 00:04–13:45 UTC, ≈ 118 by evening
   A tool for fully automatic removal of censorship (safety alignment) from transformer-based language models using directional ablation and Optuna-based parameter optimization.
   Full card: https://gitnova.dev/en/r/p-e-w/heretic.md
17. **unslothai/unsloth** — 3.4 · Steady · Language models · Python · +32 stars measured 2026-10-29, 00:05–13:45 UTC, ≈ 60 by evening
   Desktop app and web UI for locally running, fine-tuning and serving LLMs, diffusion, embedding and audio models, with OpenAI-compatible API and integrations for Claude Code, Codex and MCP agents.
   Full card: https://gitnova.dev/en/r/unslothai/unsloth.md
18. **ggml-org/llama.cpp** — 3.3 · Steady · Language models · C++ · +65 stars measured 2026-10-29, 00:04–13:45 UTC, ≈ 115 by evening
   LLM and VLM inference implemented in C/C++ with no dependencies, running on CPUs and GPUs across many hardware backends. Enables local model execution with minimal setup and quantization.
   Full card: https://gitnova.dev/en/r/ggml-org/llama.cpp.md
19. **ollama/ollama** — 3.3 · Steady · Language models · Go · +61 stars measured 2026-10-29, 00:04–13:45 UTC, ≈ 108 by evening
   Run open language models (Kimi, GLM, DeepSeek, Qwen, Gemma and more) locally via CLI and REST API with integrations into editors and assistants.
   Full card: https://gitnova.dev/en/r/ollama/ollama.md
20. **solidsf-inc/solidcv** — 3.3 · Breakout · Language models · Rust · +18 stars measured 2026-10-29, 00:05–13:45 UTC, ≈ 44 by evening
   A pure-Rust OpenCV replacement with zero FFI: classical 2D/3D vision, stereo reconstruction, and mechanistic interpretability for LLMs/VLMs with hardware kernels for Apple Silicon and NVIDIA.
   Full card: https://gitnova.dev/en/r/solidsf-inc/solidcv.md
21. **trymirai/uzu** — 3.3 · Breakout · Language models · Rust · +24 stars measured 2026-10-29, 00:03–13:45 UTC, ≈ 50 by evening
   A high-performance AI model inference engine written in Rust with APIs for Rust, Python, Swift, and TypeScript. It lets you run LLMs locally inside your app with zero latency and no inference costs.
   Full card: https://gitnova.dev/en/r/trymirai/uzu.md
22. **vllm-project/vllm** — 3.3 · Steady · Language models · Python · +62 stars measured 2026-10-29, 00:05–13:45 UTC, ≈ 104 by evening
   A high-throughput and memory-efficient inference and serving engine for large language models, using PagedAttention for KV-cache management. Supports 200+ model architectures and an OpenAI-compatible API.
   Full card: https://gitnova.dev/en/r/vllm-project/vllm.md
23. **Mesh-LLM/mesh-llm** — 3.2 · Breakout · Language models · Rust · +7 stars measured 2026-10-29, 10:52–13:45 UTC, ≈ 18 by evening
   Decentralized LLM serving system that pools GPUs and memory across machines and exposes them as one OpenAI-compatible API at localhost:9337. A model can run locally, route to a peer, or be split into layer stages across nodes.
   Full card: https://gitnova.dev/en/r/Mesh-LLM/mesh-llm.md
24. **infiniflow/ragflow** — 3.1 · Steady · Language models · Go · +56 stars measured 2026-10-29, 00:06–13:46 UTC, ≈ 90 by evening
   Open-source RAG engine that fuses document context extraction with agent capabilities to build production LLM systems. Supports parsing PDF, DOCX, spreadsheets, images and syncing from Confluence, Notion, S3.
   Full card: https://gitnova.dev/en/r/infiniflow/ragflow.md
25. **magnitudedev/magnitude** — 3.1 · Breakout · Language models · Rust · +16 stars measured 2026-10-29, 10:50–13:46 UTC, ≈ 45 by evening
   Open source inference engine for consumer hardware that profiles your machine, recommends the best local models, then downloads, tunes, and runs them, with one-click connection to AI agents.
   Full card: https://gitnova.dev/en/r/magnitudedev/magnitude.md
26. **headroomlabs-ai/headroom** — 3.0 · Steady · Language models · Python · +64 stars measured 2026-10-29, 00:05–13:46 UTC, ≈ 112 by evening
   A local library, proxy and MCP server that compress tool outputs, logs, files and RAG chunks before they reach the LLM, cutting token usage without changing answers.
   Full card: https://gitnova.dev/en/r/headroomlabs-ai/headroom.md
27. **slee-persis/GVS5H** — 2.9 · Steady · Language models · Python · +15 stars measured 2026-10-29, 00:05–13:46 UTC, ≈ 25 by evening
   A training-free method that runs five instances of one LLM in parallel, coordinating through a shared filesystem to solve hard coding problems, evaluated on LiveCodeBench.
   Full card: https://gitnova.dev/en/r/slee-persis/GVS5H.md
28. **Tencent-Hunyuan/Hy-SimulMT** — 2.8 · Early signal · Language models · Python · +9 stars measured 2026-10-29, 00:03–13:46 UTC, ≈ 16 by evening
   Tencent Hunyuan's simultaneous speech translation model for real-time bidirectional Chinese-English translation with voice cloning. The repository provides the SimulInfer inference service and SimulTrain training tools.
   Full card: https://gitnova.dev/en/r/Tencent-Hunyuan/Hy-SimulMT.md
29. **langfuse/langfuse** — 2.6 · Steady · Language models · TypeScript · +38 stars measured 2026-10-29, 00:06–13:47 UTC, ≈ 62 by evening
   Open source LLM engineering platform for tracing, evaluating, and debugging AI applications, with self-hosting and prompt management.
   Full card: https://gitnova.dev/en/r/langfuse/langfuse.md
30. **xai-org/x-algorithm** — 2.6 · Breakout · Language models · Rust · +18 stars measured 2026-10-29, 00:06–13:47 UTC, ≈ 33 by evening
   Open-source code of X's For You feed algorithm: it blends in-network posts with ML-retrieved out-of-network content, filters them, and ranks posts with a transformer model. Published for transparency and study of how the feed is built.
   Full card: https://gitnova.dev/en/r/xai-org/x-algorithm.md
31. **mw00/project-maya** — 2.6 · Steady · Language models · Cuda · +26 stars measured 2026-10-29, 00:05–13:47 UTC, ≈ 48 by evening
   Local engine, server and dashboard to run the 321B MoE model GLM-5.3-Flash on one or two NVIDIA GPUs, offloading experts to RAM and NVMe.
   Full card: https://gitnova.dev/en/r/mw00/project-maya.md
32. **jaredpalmer/kev** — 2.5 · Steady · Language models · Python · +20 stars measured 2026-10-29, 00:06–13:47 UTC, ≈ 35 by evening
   kev is a LoRA adapter with a small readout head on top of Qwen2.5-0.5B that answers many typed questions about a document in a single forward pass, returning calibrated probabilities instead of text.
   Full card: https://gitnova.dev/en/r/jaredpalmer/kev.md
33. **promptfoo/promptfoo** — 2.5 · Steady · Language models · TypeScript · +23 stars measured 2026-10-29, 00:06–13:47 UTC, ≈ 40 by evening
   CLI and library for evaluating and red-teaming LLM apps: automated prompt evals, model comparison, and vulnerability scanning. Runs locally and integrates with CI/CD.
   Full card: https://gitnova.dev/en/r/promptfoo/promptfoo.md
34. **tensorflow/tensorflow** — 2.5 · Steady · Language models · C++ · +10 stars measured 2026-10-29, 00:07–13:48 UTC, ≈ 19 by evening
   An open source machine learning platform from Google Brain with Python and C++ APIs for training and deploying neural networks. Used by researchers and developers of ML applications.
   Full card: https://gitnova.dev/en/r/tensorflow/tensorflow.md
35. **firecrawl/anydoc** — 2.3 · Steady · Language models · Rust · +29 stars measured 2026-10-29, 00:07–13:48 UTC, ≈ 48 by evening
   Rust library that converts office documents (Word, PowerPoint, Excel, OpenDocument, RTF, EPUB, CSV, PDF) into clean Markdown, with Node.js, Python and browser bindings. Built to make documents LLM-ready.
   Full card: https://gitnova.dev/en/r/firecrawl/anydoc.md
36. **deeplethe/utopia** — 2.3 · Steady · Language models · Rust · +19 stars measured 2026-10-29, 00:08–13:49 UTC, ≈ 32 by evening
   Utopia is an open-source enterprise knowledge platform: a bitemporal knowledge graph, RAG search and agent workflows on a single Rust binary plus PostgreSQL. It deploys offline, for companies that need a controlled knowledge foundation…
   Full card: https://gitnova.dev/en/r/deeplethe/utopia.md
37. **kvcache-ai/Mooncake** — 2.3 · Steady · Language models · C++ · +13 stars measured 2026-10-29, 00:08–13:49 UTC, ≈ 20 by evening
   LLM serving platform built on a KVCache-centric disaggregated architecture: separates prefill and decode and transfers KV cache between nodes over RDMA. Powers Kimi in production at Moonshot AI.
   Full card: https://gitnova.dev/en/r/kvcache-ai/Mooncake.md
38. **Albert-Weasker/niubigeo** — 2.3 · Steady · Language models · TypeScript · +22 stars measured 2026-10-29, 00:13–13:48 UTC, ≈ 44 by evening
   Open-source tool for tracking brand visibility and competitors in AI answers: given a domain, it shows how models describe your product, whom they recommend, and which sources they cite.
   Full card: https://gitnova.dev/en/r/Albert-Weasker/niubigeo.md
39. **MemTensor/MemT-1.0** — 2.1 · Steady · Language models · Python · +13 stars measured 2026-10-29, 00:09–13:49 UTC, ≈ 23 by evening
   A family of memory foundation models with native long-term memory: dynamic local memory and trainable hyper memory retain history without accessing the original context, cutting storage roughly 66x versus full-context KV cache at 128K.
   Full card: https://gitnova.dev/en/r/MemTensor/MemT-1.0.md
40. **tesseract-ocr/tesseract** — 2.1 · Steady · Language models · C++ · +11 stars measured 2026-10-29, 00:08–13:50 UTC, ≈ 19 by evening
   Open-source OCR engine libtesseract and CLI tool tesseract for recognizing text in images. Supports 100+ languages, PNG/JPEG/TIFF input, and output in text, hOCR, PDF, TSV, ALTO.
   Full card: https://gitnova.dev/en/r/tesseract-ocr/tesseract.md
41. **mistralai/timestral** — 2.1 · Cooling · Language models · Python · +15 stars measured 2026-10-29, 00:08–13:50 UTC, ≈ 26 by evening
   A family of pretrained foundation models for multivariate probabilistic time series forecasting with covariate support. Enables zero-shot forecasting on CPU or GPU without training.
   Full card: https://gitnova.dev/en/r/mistralai/timestral.md
42. **vllm-project/semantic-router** — 2.0 · Steady · Language models · Go · +3 stars measured 2026-10-29, 10:53–13:50 UTC, ≈ 8 by evening
   A programmable routing layer for Mixture-of-Models systems across heterogeneous LLM infrastructure: it selects or composes the right model path per request based on request signals, user preferences, and application policies. It helps…
   Full card: https://gitnova.dev/en/r/vllm-project/semantic-router.md
43. **mudler/LocalAI** — 2.0 · Steady · Language models · Go · +15 stars measured 2026-10-29, 00:08–13:50 UTC, ≈ 26 by evening
   LocalAI is an open-source AI engine that runs LLMs, vision, voice, image and video models on any hardware, including CPU-only, behind an OpenAI-compatible API.
   Full card: https://gitnova.dev/en/r/mudler/LocalAI.md
44. **NVIDIA/cutlass** — 1.9 · Steady · Language models · C++ · +7 stars measured 2026-10-29, 00:10–13:50 UTC, ≈ 11 by evening
   CUTLASS is a collection of CUDA C++ templates and Python DSLs (CuTe DSL) for implementing high-performance linear algebra, primarily GEMM, on NVIDIA GPUs. It targets researchers and performance engineers writing optimized Tensor Core…
   Full card: https://gitnova.dev/en/r/NVIDIA/cutlass.md
45. **VectifyAI/PageIndex** — 1.9 · Cooling · Language models · Python · +8 stars measured 2026-10-29, 00:07–13:50 UTC, ≈ 18 by evening
   A vectorless RAG engine that builds a hierarchical tree index of a document and retrieves by LLM reasoning over that tree, the way a human navigates a report. Aimed at long professional PDFs such as financial, legal and technical documents.
   Full card: https://gitnova.dev/en/r/VectifyAI/PageIndex.md
46. **google-research/kauldron** — 1.8 · Steady · Language models · Python · +2 stars measured 2026-10-29, 00:05–13:50 UTC, ≈ 4 by evening
   A modular Google Research library for training ML models on JAX, focused on research velocity and flexible configuration.
   Full card: https://gitnova.dev/en/r/google-research/kauldron.md
47. **microsoft/BitNet** — 1.8 · Steady · Language models · C++ · +7 stars measured 2026-10-29, 00:09–13:51 UTC, ≈ 13 by evening
   Official inference framework for 1-bit LLMs (BitNet b1.58) with optimized kernels for CPU and GPU. Runs ternary models fast and losslessly, including a 100B model on a single CPU.
   Full card: https://gitnova.dev/en/r/microsoft/BitNet.md
48. **FareedKhan-dev/kimi-k3-in-c** — 1.8 · Steady · Language models · C · +7 stars measured 2026-10-29, 00:09–13:51 UTC, ≈ 13 by evening
   A portable C99 inference engine that runs the 2.78-trillion-parameter Kimi K3 model on a single CPU within 8.24 GB of RAM, with no BLAS, framework, or GPU. The model streams from disk, and memory only affects speed.
   Full card: https://gitnova.dev/en/r/FareedKhan-dev/kimi-k3-in-c.md
49. **cocoindex-io/cocoindex** — 1.8 · Steady · Language models · Rust · +7 stars measured 2026-10-29, 00:10–13:51 UTC, ≈ 12 by evening
   Rust-based incremental engine that turns codebases, PDFs, Slack and notes into continuously fresh context for AI agents and RAG apps, recomputing only the delta.
   Full card: https://gitnova.dev/en/r/cocoindex-io/cocoindex.md
50. **google-pytorch/torch_tpu** — 1.7 · Cooling · Language models · C++ · +3 stars measured 2026-10-29, 00:07–13:51 UTC, ≈ 6 by evening
   TorchTPU is a PyTorch backend for Google TPUs that lets you run PyTorch models and operations on TPU devices, with custom ATen kernels, an XLA compilation cache, and distributed training support.
   Full card: https://gitnova.dev/en/r/google-pytorch/torch_tpu.md
51. **chenglou/handwired** — 1.7 · Cooling · Language models · TypeScript · +6 stars measured 2026-10-29, 00:09–13:51 UTC, ≈ 12 by evening
   Turns regular JavaScript code into a differentiable neural net: if-else, loops and state are compiled into matrix multiplications and ReLU with no training. Runs on JS or WebGPU.
   Full card: https://gitnova.dev/en/r/chenglou/handwired.md
52. **mostlygeek/llama-swap** — 1.7 · Steady · Language models · Go · +5 stars measured 2026-10-29, 00:10–13:52 UTC, ≈ 10 by evening
   A Go proxy server that runs multiple local models and hot-swaps between them on demand, compatible with OpenAI and Anthropic APIs. Works with llama.cpp, vllm, stable-diffusion.cpp and other backends.
   Full card: https://gitnova.dev/en/r/mostlygeek/llama-swap.md
53. **FlashML-org/FreeToken** — 1.6 · Steady · Language models · Python · +3 stars measured 2026-10-29, 00:09–13:52 UTC, ≈ 7 by evening
   A Mixture-of-Experts inference engine for running frontier-scale open-weight MoE models (290B+) on consumer hardware — GPUs, CPUs, and host memory. Aimed at users who want to run frontier models locally.
   Full card: https://gitnova.dev/en/r/FlashML-org/FreeToken.md
54. **cobanov/PS5LM** — 1.6 · Cooling · Language models · C++ · +3 stars measured 2026-10-29, 00:09–13:52 UTC, ≈ 6 by evening
   A llama.cpp port for jailbroken PlayStation 5 that runs GGUF models natively on the console via a payload, with a browser-based model library and chat. Only small models up to 1.6 GB work so far.
   Full card: https://gitnova.dev/en/r/cobanov/PS5LM.md
55. **NVIDIA/spark-local-inference** — 1.5 · Cooling · Language models · Rust · +2 stars measured 2026-10-29, 00:09–13:52 UTC, ≈ 4 by evening
   Flare is a local inference server for NVIDIA SM120/SM121 GPUs aimed at long conversations, coding agents and repeated tool use. It compiles models with PyTorch and serves them from a Rust runtime to use workstation memory and compute more…
   Full card: https://gitnova.dev/en/r/NVIDIA/spark-local-inference.md
56. **microsoft/onnxruntime** — 1.5 · Steady · Language models · C++ · +0 stars measured 2026-10-29, 00:09–13:52 UTC, ≈ 1 by evening
   Cross-platform accelerator for ML inference and training. Runs models from PyTorch, TensorFlow, scikit-learn and others via the ONNX format with graph optimizations and hardware acceleration.
   Full card: https://gitnova.dev/en/r/microsoft/onnxruntime.md
57. **MiaAI-Lab/DeepSeek-v4.1-Flash-1017-2x-DGX-Sparks** — 1.5 · Cooling · Language models · Python · +7 stars measured 2026-10-29, 00:10–13:53 UTC, ≈ 12 by evening
   Scripts to serve the quantized DeepSeek-V4.1-Flash-1017 (EXL3, 2.9 bpw) across two NVIDIA DGX Sparks via the TensorFold engine, exposing an OpenAI-compatible API with up to a 1M-token window and DSpark speculative decoding.
   Full card: https://gitnova.dev/en/r/MiaAI-Lab/DeepSeek-v4.1-Flash-1017-2x-DGX-Sparks.md
58. **deepseek-ai/FlashMLA** — 1.4 · Steady · Language models · C++ · +4 stars measured 2026-10-29, 00:09–13:52 UTC, ≈ 7 by evening
   DeepSeek's library of optimized attention kernels (sparse and dense) for running DeepSeek-V4.1 inference on NVIDIA GPUs and Huawei Ascend NPUs, with FP8/FP4 KV cache support.
   Full card: https://gitnova.dev/en/r/deepseek-ai/FlashMLA.md
59. **nace-ai/kapitsa** — 1.4 · Cooling · Language models · C++ · +4 stars measured 2026-10-29, 00:08–13:53 UTC, ≈ 8 by evening
   Apple Silicon optimized inference engine for Qwen and Nemotron models with OpenAI- and Anthropic-compatible APIs. Runs local LLMs on Mac via CLI and a web chat.
   Full card: https://gitnova.dev/en/r/nace-ai/kapitsa.md
60. **drumih/turbo-fieldfare** — 1.4 · Steady · Language models · Swift · +6 stars measured 2026-10-29, 00:10–13:53 UTC, ≈ 10 by evening
   A custom Swift + Metal runtime for running Gemma 4 26B-A4B on Apple Silicon Macs: keeps ~2 GB in memory and streams experts from SSD, so it works even on 8 GB RAM machines.
   Full card: https://gitnova.dev/en/r/drumih/turbo-fieldfare.md
61. **martina9516/uncensored-ai-models** — 1.3 · Cooling · Language models · +0 stars measured 2026-10-29, 00:10–13:53 UTC, ≈ 0 by evening
   A single-key API gateway to Claude, GPT, Gemini, Grok, DeepSeek, GLM and Qwen, including uncensored and abliterated variants. OpenAI- and Anthropic-compatible, pay per token.
   Full card: https://gitnova.dev/en/r/martina9516/uncensored-ai-models.md
62. **MoonshotAI/Kimi-K3** — 1.3 · Steady · Language models · +4 stars measured 2026-10-29, 00:11–13:53 UTC, ≈ 7 by evening
   Kimi K3 is an open-weight native multimodal agentic MoE language model with 2.8T parameters and a 1M-token context window, understanding text, images and video. It targets long-horizon coding, agentic knowledge work and complex reasoning.
   Full card: https://gitnova.dev/en/r/MoonshotAI/Kimi-K3.md
63. **ml-explore/mlx** — 1.3 · Steady · Language models · C++ · +2 stars measured 2026-10-29, 00:08–13:54 UTC, ≈ 4 by evening
   MLX is an array and machine learning framework from Apple, optimized for Apple silicon with unified memory and lazy computation. Its Python, C++, C, and Swift APIs mirror NumPy and PyTorch, simplifying model training and deployment.
   Full card: https://gitnova.dev/en/r/ml-explore/mlx.md
64. **eagle-hq/EAGLE-4** — 1.3 · Cooling · Language models · Python · +2 stars measured 2026-10-29, 00:10–13:53 UTC, ≈ 4 by evening
   Official inference harness for EAGLE-4 speculative decoding, adding a flexible drafter plugin to SGLang, vLLM and TokenSpeed that adapts to batch size, hardware and workload. It speeds up serving of LLMs such as Qwen3.8-27B while…
   Full card: https://gitnova.dev/en/r/eagle-hq/EAGLE-4.md
65. **openvinotoolkit/openvino** — 1.2 · Steady · Language models · C++ · +3 stars measured 2026-10-29, 00:10–13:54 UTC, ≈ 5 by evening
   OpenVINO is an open-source toolkit for optimizing and deploying deep learning model inference on Intel CPU, GPU and NPU. It supports models from PyTorch, TensorFlow, ONNX, Keras, PaddlePaddle, JAX/Flax and Hugging Face with C++, Python, C…
   Full card: https://gitnova.dev/en/r/openvinotoolkit/openvino.md
66. **recursive-oss/rec** — 1.2 · Cooling · Language models · Python · +5 stars measured 2026-10-29, 00:10–13:54 UTC, ≈ 8 by evening
   A local inference engine for Recursive's language models, optimized for Apple Silicon and faster than llama.cpp. It runs the model on your machine and exposes an OpenAI-compatible API for clients.
   Full card: https://gitnova.dev/en/r/recursive-oss/rec.md
67. **lemonade-sdk/lemonade** — 1.2 · Steady · Language models · C++ · +0 stars measured 2026-10-29, 10:51–13:54 UTC, ≈ 2 by evening
   A local AI server that runs optimized LLMs, whisper, TTS, and image generation on the user's GPU and NPU, exposing OpenAI-, Anthropic-, and Ollama-compatible APIs. Available as a service or an embeddable binary for apps.
   Full card: https://gitnova.dev/en/r/lemonade-sdk/lemonade.md
68. **astroautomata/PySR** — 1.2 · Cooling · Language models · Python · +4 stars measured 2026-10-29, 00:10–13:55 UTC, ≈ 6 by evening
   PySR is a symbolic regression library for Python and Julia that searches for interpretable analytic expressions fitting a dataset, instead of a black-box neural network.
   Full card: https://gitnova.dev/en/r/astroautomata/PySR.md
69. **deepseek-ai/DeepSpec** — 1.1 · Steady · Language models · Python · +5 stars measured 2026-10-29, 00:11–13:54 UTC, ≈ 7 by evening
   Full-stack pipeline for training and evaluating draft models for speculative decoding: data preparation, training against a target-model cache, and acceptance measurement on benchmarks. Supports DSpark, DFlash and Eagle3.
   Full card: https://gitnova.dev/en/r/deepseek-ai/DeepSpec.md
70. **byenzyme/enzyme** — 1.1 · Steady · Language models · Rust · +0 stars measured 2026-10-29, 00:08–13:55 UTC, ≈ 1 by evening
   Local-first indexing tool that turns Markdown notes into a searchable knowledge base for coding agents, ranking passages by wording and meaning on your own machine.
   Full card: https://gitnova.dev/en/r/byenzyme/enzyme.md
71. **mizorewww/laya-mlx** — 1.0 · Steady · Language models · Python · +1 star measured 2026-10-29, 00:11–13:55 UTC, ≈ 2 by evening
   Native MLX runtime for Laya typed decision models: returns probabilities for choices, scores or truth values without text generation, running locally on Apple Silicon.
   Full card: https://gitnova.dev/en/r/mizorewww/laya-mlx.md
72. **ValerioDolci/capy-flash** — 0.9 · Steady · Language models · C++ · +3 stars measured 2026-10-29, 00:11–13:55 UTC, ≈ 5 by evening
   A C++ inference engine that runs the large Qwen3.8-Flash-Next MoE model (125B) in NVFP4 on two 16 GB consumer GPUs: routed experts are offloaded to host RAM and computed on the CPU, the rest runs tensor-parallel. It also serves the dense…
   Full card: https://gitnova.dev/en/r/ValerioDolci/capy-flash.md
73. **h563k/chat-keeper** — 0.9 · Cooling · Language models · CSS · +0 stars measured 2026-10-29, 00:11–13:55 UTC, ≈ 2 by evening
   A third-party mirror of the Claude web interface plus API access to Claude, Gemini and OpenAI, reachable directly from China without a VPN. The repository mainly serves as a landing page and entry point for the paid service.
   Full card: https://gitnova.dev/en/r/h563k/chat-keeper.md
74. **BaruqueRodrigues/llm-legislativo** — 0.8 · Cooling · Language models · Python · +0 stars measured 2026-10-29, 00:11–13:55 UTC, ≈ 0 by evening
   Large-scale Brazilian legislative corpus and a causal Transformer language model trained from scratch in pure PyTorch without pretrained weights or closed APIs.
   Full card: https://gitnova.dev/en/r/BaruqueRodrigues/llm-legislativo.md
75. **apple-aiml-research/ml-rsi-inference-bench** — 0.8 · Steady · Language models · Python · +1 star measured 2026-10-29, 00:11–13:55 UTC, ≈ 2 by evening
   A benchmark for evaluating coding agents that optimize and repair the LLM inference stack in real engines like vLLM and SGLang on real GPUs. Agents work on the live engine and are graded by running what they built.
   Full card: https://gitnova.dev/en/r/apple-aiml-research/ml-rsi-inference-bench.md
76. **justxor/deepseek-local-guide-ru** — 0.8 · Steady · Language models · Shell · +2 stars measured 2026-10-29, 00:11–13:56 UTC, ≈ 3 by evening
   Russian-language guide to running DeepSeek models (R1, V4) locally on PC, Mac, and server: Ollama, llama.cpp, LM Studio, vLLM, Open WebUI, API, RAG, agents, and fine-tuning.
   Full card: https://gitnova.dev/en/r/justxor/deepseek-local-guide-ru.md
77. **deepopen-com/deepopen** — 0.6 · Cooling · Language models · Python · +5 stars measured 2026-10-29, 00:12–13:56 UTC, ≈ 7 by evening · unusual star pattern (heuristic)
   Open-source non-autoregressive System 1 decision engine for typed decisions: returns typed answers (choice, score, noul) in a single forward pass across 100+ languages without generating text. Suited for classification, routing and…
   Full card: https://gitnova.dev/en/r/deepopen-com/deepopen.md
78. **KohakuBlueleaf/KohakuMossTrainer** — 0.6 · Steady · Language models · Python · +0 stars measured 2026-10-29, 00:11–13:56 UTC, ≈ 0 by evening
   A finetuning framework for MOSS-Image: every LyCORIS algorithm and full finetune on text-image data at native resolutions, plus inference, on Windows and Linux from 6 GB cards to multi-GPU nodes.
   Full card: https://gitnova.dev/en/r/KohakuBlueleaf/KohakuMossTrainer.md
79. **Taichu-AI/ZDTaichu5.0-9B** — 0.5 · Steady · Language models · Python · +1 star measured 2026-10-29, 00:12–13:56 UTC, ≈ 2 by evening · unusual star pattern (heuristic)
   ZDTaichu5.0-9B is a multimodal foundation model built on Qwen3.5-9B and C-RADIOv4-H for image, video, spatial and embodied reasoning, and agentic tasks.
   Full card: https://gitnova.dev/en/r/Taichu-AI/ZDTaichu5.0-9B.md
80. **strot-ai/STROT-Compiler** — 0.4 · Steady · Language models · Python · +0 stars measured 2026-10-29, 00:12–13:56 UTC, ≈ 0 by evening
   STROT compiles a natural-language decision intent into a frozen local Python artifact that classifies, scores, or routes records without calling an LLM at runtime. It targets teams wanting cheap, deterministic, type-safe decisions on…
   Full card: https://gitnova.dev/en/r/strot-ai/STROT-Compiler.md
81. **heterodoxin/AbliteratedOutTheBox** — 0.4 · Cooling · Language models · C++ · +0 stars measured 2026-10-29, 00:11–13:56 UTC, ≈ 0 by evening
   A llama.cpp fork that performs abliteration at inference time: it reads refusal directions from the prompt itself and removes them inside the graph without touching the weights on disk. Runs any supported model with no fine-tuning and no…
   Full card: https://gitnova.dev/en/r/heterodoxin/AbliteratedOutTheBox.md
82. **NineAbyss/OR-License** — 0.2 · Quiet · Language models · Python · +0 stars measured 2026-10-29, 00:12–13:56 UTC, ≈ 0 by evening
   A 31-task benchmark evaluating LLM agents as operations research practitioners: the agent requests information, writes an optimization program in an isolated environment, and submits solution files.
   Full card: https://gitnova.dev/en/r/NineAbyss/OR-License.md
83. **prabanta-dev/janas** — 0.2 · Steady · Language models · C · +0 stars measured 2026-10-29, 00:12–13:56 UTC, ≈ 0 by evening
   Janas is a family of AI tools in C with no dependencies. Janas-LLM runs large MoE models (up to 80B, 48 GB) on a 32 GB laptop by streaming experts from NVMe SSD, and outperforms llama.cpp on all 13 tested models.
   Full card: https://gitnova.dev/en/r/prabanta-dev/janas.md
84. **composo-ai/DishwasherBench** — 0.1 · Quiet · Language models · TypeScript · +0 stars measured 2026-10-29, 00:12–13:56 UTC, ≈ 0 by evening
   A browser game about loading a dishwasher, built on a physical simulation of a real Bosch machine to benchmark language models. Players load dishes, run the Eco 50 programme and score points for items that come out clean.
   Full card: https://gitnova.dev/en/r/composo-ai/DishwasherBench.md
85. **7API/glm-5.3-abliterated** — 0.1 · Cooling · Language models · +0 stars measured 2026-10-29, 00:12–13:56 UTC, ≈ 0 by evening
   A guide and prompt collection for GLM 5.3 abliterated/uncensored models, with 35 jailbreak prompts and access through a single OpenAI- and Anthropic-compatible 7API key.
   Full card: https://gitnova.dev/en/r/7API/glm-5.3-abliterated.md
86. **rockmatrix-ai/EdgeRock** — 0.0 · Quiet · Language models · Python · Not enough measurements today
   YANchor-4B is a 4.58B-parameter general-purpose recurrent language model that preserves crucial information as memory anchors, enabling long-horizon reasoning in O(N) time and O(1) memory. The repo provides inference and evaluation code.
   Full card: https://gitnova.dev/en/r/rockmatrix-ai/EdgeRock.md
87. **cdcailf/cjkfold** — 0.0 · Steady · Language models · Python · Not enough measurements today
   Python library that folds look-alike CJK glyphs (Kangxi radicals and CJK radical supplements) back to the Han characters they render as, so text extracted from PDFs matches in search and comparison. Zero-dependency, pure functions.
   Full card: https://gitnova.dev/en/r/cdcailf/cjkfold.md
88. **Augmented-Nature/BioOntology-MCP-Server** — 0.0 · Quiet · Language models · JavaScript · Not enough measurements today
   Unofficial MCP server exposing the BioOntology API for searching, annotating, and exploring 1,200+ biological ontologies. It lets MCP-compatible clients like Claude Desktop work with biomedical ontology data.
   Full card: https://gitnova.dev/en/r/Augmented-Nature/BioOntology-MCP-Server.md
89. **7API/uncensored-ai-api** — 0.0 · Cooling · Language models · +0 stars measured 2026-10-29, 00:12–12:55 UTC, ≈ 0 by evening · unusual star pattern (heuristic)
   A gateway giving one API key to Claude, GPT, Gemini, Grok, DeepSeek, GLM and Kimi, including uncensored and abliterated variants. OpenAI- and Anthropic-compatible, pay per token.
   Full card: https://gitnova.dev/en/r/7API/uncensored-ai-api.md
90. **JackWang666666/my-rag** — 0.0 · Steady · Language models · Java · Not enough measurements today
   Retrieval-augmented generation knowledge base system with a Next.js frontend and two equivalent backends (FastAPI/LangChain and Spring Boot/Spring AI) sharing the same MySQL, Milvus and Neo4j data.
   Full card: https://gitnova.dev/en/r/JackWang666666/my-rag.md
91. **routerhubSI/uncensored-ai-api** — 0.0 · Cooling · Language models · +0 stars measured 2026-10-29, 00:13–12:23 UTC, ≈ 0 by evening
   Gateway to AI APIs and models (Claude, GPT, Gemini, Grok, DeepSeek, GLM, Qwen) with one key, OpenAI- and Anthropic-compatible, offering uncensored and abliterated variants with pay-per-token billing.
   Full card: https://gitnova.dev/en/r/routerhubSI/uncensored-ai-api.md

For agents: https://gitnova.dev/llms.txt · search: https://gitnova.dev/api/v1/search?q=… · MCP: https://gitnova.dev/mcp · any page as Markdown — append .md to its URL