Topics / #reasoning

Reasoning projects on GitHub

Approaches and architectures for improving model reasoning capabilities.

As of , the most-starred Reasoning projects on GitHub are DeepSeek-V3 (104.5K stars), DeepSeek-R1 (91.9K stars) and dive-into-llms (56.8K stars). GitHub.Locker tracks 46 Reasoning repositories with 584.2K stars combined. The fastest-growing on October 16 was StartLux-Decision, with +150 stars in a day.

46Projects
584.2KStars in total
+609Stars in a day
Browse in the catalog →github.com/topics/reasoning ↗

Hot right now

Most stars gained in a day · Oct 16

+150

StartLux-Decision

StartLuxLabs/StartLux-Decision · ★ 552

StartLux-Decision is a family of typed decision models (five dense sizes from 0.8B to 27B plus a 35B-A3B mixture-of-experts) that answer a question with a probability for each option instead of generating text.

+87

math

openai/math · ★ 15K

A public catalogue of 719 mathematical manuscripts grouped into 372 families, written by an internal OpenAI model during evaluations on open research problems. About 42% of top-line results already have Lean formalizations.

+75

PageIndex

VectifyAI/PageIndex · ★ 39.2K

PageIndex is a vectorless, reasoning-based RAG (Retrieval-Augmented Generation) engine. Instead of a vector database and chunking, it builds a hierarchical tree index that mirrors a document's structure and lets an LLM reason its way through that tree to find relevant sections, the way a human expert flips to the right part of a report.

+63

AnyJev

nokia-applied-research/AnyJev · ★ 1.4K

A Python library that turns any open LLM into a Jev-style decision model: ask a typed question, get back a choice with a probability you can threshold — no generation, no parsing, no fine-tuning.

The all-time top

46 projects · by stars

  1. #1 · deepseek-ai/DeepSeek-V3

    DeepSeek-V3

    DeepSeek-V3 is an open-weight Mixture-of-Experts (MoE) language model with 671B total parameters but only 37B activated per token, built for efficient yet high-performing inference and matching leading closed-source models on many benchmarks.

    104.5K
  2. #2 · deepseek-ai/DeepSeek-R1

    DeepSeek-R1

    DeepSeek-R1 is an open-source family of reasoning-focused large language models from DeepSeek, trained with large-scale reinforcement learning to handle math, code, and general reasoning at a level comparable to OpenAI's o1.

    91.9K
  3. #3 · Lordog/dive-into-llms

    dive-into-llms

    It's a free Chinese-language tutorial series called 'Dive into LLMs' (《动手学大模型》) that teaches large language model concepts through hands-on coding exercises, built from Shanghai Jiao Tong University's NLP and AI-security course materials.

    56.8K
  1. 4

    PageIndex

    39.2KVectifyAI/PageIndex · Python

    PageIndex is a vectorless, reasoning-based RAG (Retrieval-Augmented Generation) engine. Instead of a vector database and chunking, it builds a hierarchical tree index that mirrors a document's structure and lets an LLM reason its way through that tree to find relevant sections, the way a human expert flips to the right part of a report.

  2. 5

    Qwen3

    27.7KQwenLM/Qwen3 · Python

    Qwen3 is a series of open-weight large language models built by the Qwen team at Alibaba Cloud, released as both dense and Mixture-of-Experts (MoE) models from 0.6B up to 235B-A22B parameters, with switchable 'thinking' and 'non-thinking' reasoning modes.

  3. 6

    open-r1

    26.5Khuggingface/open-r1 · Python

    Open R1 is a fully open effort to reproduce the training pipeline behind DeepSeek-R1, providing scripts to run SFT and GRPO training, generate synthetic reasoning data, and evaluate the resulting reasoning models.

  4. 7

    gpt-oss

    20.5Kopenai/gpt-oss · Python

    The gpt-oss series: two open-weight language models from OpenAI — gpt-oss-120b and gpt-oss-20b — built for powerful reasoning, agentic tasks, and versatile developer use cases. The repo itself adds reference implementations, tools, and example clients around the models.

  5. 8

    Qwen3-VL

    20.1KQwenLM/Qwen3-VL · Jupyter Notebook

    Qwen3-VL is a family of multimodal vision-language models built by the Qwen team at Alibaba Cloud, combining strong text understanding with deep visual perception, reasoning, and agent capabilities.

  6. 9

    Thinking-Claude

    17Krichards199999/Thinking-Claude · TypeScript

    Thinking Claude is a project with two parts: a 'Thinking Protocol' - custom instructions that make Claude reason more deeply and systematically before answering - and a browser extension that renders that thinking process as neat, collapsible sections in the Claude.ai chat UI.

  7. 10

    math

    15Kopenai/math · Lean

    A public catalogue of 719 mathematical manuscripts grouped into 372 families, written by an internal OpenAI model during evaluations on open research problems. About 42% of top-line results already have Lean formalizations.

  8. 11

    semantica

    14.1Ksemantica-agi/semantica · Python

    Semantica is an open-source, graph-native infrastructure layer that ingests enterprise data and builds a structured Context Graph and knowledge graph with deterministic reasoning, ontology governance, and full decision provenance, so AI systems can be explained and audited instead of treated as black boxes.

  9. 12

    TinyZero

    13.2KJiayi-Pan/TinyZero · Python

    A minimal reproduction of DeepSeek R1-Zero: through reinforcement learning on countdown and multiplication tasks, a 3B base LM develops self-verification and search abilities all on its own.

  10. 13

    HRM

    12.7Ksapientinc/HRM · Python

    The official release of the Hierarchical Reasoning Model (HRM): a brain-inspired recurrent architecture that reaches significant computational depth with only 27 million parameters, trained on ~1000 examples without pretraining or chain-of-thought data.

  11. 14

    LLMSurvey

    12.2KRUCAIBox/LLMSurvey · Python

    It is the official GitHub page for the survey paper 'A Survey of Large Language Models' — a curated collection of papers and resources on how LLMs are built, adapted, and used.

  12. 15

    MiniCPM

    11.6KOpenBMB/MiniCPM · Jupyter Notebook

    MiniCPM is OpenBMB's family of small, open-source LLMs built for on-device and resource-constrained use. The newest release, MiniCPM5-2B, is a dense 2B-parameter model that reaches open-source state-of-the-art for its size class while staying competitive with much larger 4B-class models.

  13. 16

    Kimi-K2

    11.1KMoonshotAI/Kimi-K2

    Kimi K2 is a large language model series from Moonshot AI: a mixture-of-experts model with 1 trillion total parameters (32B activated per token), trained for frontier knowledge, reasoning, coding and especially agentic tool-use tasks.

  14. 17

    arc-task-gen

    11Kpathwaycom/arc-task-gen · Python

    A generator for fresh, original ARC-AGI-1-style puzzle tasks whose statistical distribution matches the public ARC-AGI-1 evaluation set, so models can be tested on problems they haven't seen before.

  15. 18

    InternVL

    10.2KOpenGVLab/InternVL · Python

    A family of open-source multimodal chat models that understand images, video and text, positioned as an open alternative to GPT-4o. InternVL2.5-78B was the first open-source model to top 70% on the MMMU benchmark.

  16. 19

    Kimi-K3

    8.9KMoonshotAI/Kimi-K3

    Kimi K3 is MoonshotAI's open-weight, native multimodal agentic model — a 2.8T-parameter Mixture-of-Experts system built for frontier-level long-horizon coding, agentic knowledge work, and reasoning.

  17. 20

    GLM-5

    7.3Kzai-org/GLM-5

    GLM-5 is a family of large open-weight language models from Z.ai built for agentic engineering and complex, long-horizon coding tasks, scaling up to 744B parameters with sparse attention for efficient long-context serving.

  18. 21

    dsh-routing-suite

    7Kyjh051108/dsh-routing-suite · JavaScript

    dsh-routing-suite is a three-part kit for the DSH agent runtime: a runtime injector, a task-aware reasoning-mode router preset, and a graded task protocol — all shipped in one repo.

  19. 22

    TinyRecursiveModels

    6.6KSamsungSAILMontreal/TinyRecursiveModels · Python

    This is TRM (Tiny Recursive Model), the code behind the paper 'Less is More: Recursive Reasoning with Tiny Networks': a tiny 7M-parameter network that scores 45% on ARC-AGI-1 and 8% on ARC-AGI-2 through recursive reasoning.

  20. 23

    ai-agents-from-scratch

    4.9Kpguso/ai-agents-from-scratch · JavaScript

    A hands-on tutorial repo that teaches you to build AI agents from first principles using local LLMs via node-llama-cpp, with no frameworks involved, so you understand function calling, memory, and ReAct before using production frameworks.

  21. 24

    council-of-high-intelligence

    4.7K0xNyk/council-of-high-intelligence · Shell

    Council of High Intelligence runs structured multi-perspective deliberation on hard decisions by assigning a question to 18 distinct analytical personas (Aristotle, Socrates, Feynman, Linus Torvalds and others), forcing independent first positions, cross-examination, and a final verdict that preserves unresolved questions, dissent, kill criteria, and a concrete next step.

  22. 25

    adhd

    4.4KUditAkhourii/adhd · TypeScript

    ADHD is a skill for coding agents that spawns several isolated, differently-framed reasoning branches in parallel, then runs a separate critic pass to score, prune, and deepen the survivors, instead of walking one shared chain of thought.

  23. 26

    bdh

    3.6Kpathwaycom/bdh · Python

    BDH (Dragon Hatchling) is a biologically inspired large language model architecture that connects deep learning with neuroscience, aiming to explain how reasoning and generalization emerge in AI systems at the level of individual neuron interactions.

  24. 27

    Attention-Residuals

    3.5KMoonshotAI/Attention-Residuals

    AttnRes (Attention Residuals) is a drop-in replacement for standard residual connections in Transformers, letting each layer use learned, input-dependent attention over all previous layer outputs instead of simple additive accumulation.

  25. 28

    MSA

    3.5KEverMind-AI/MSA · Python

    MSA (Memory Sparse Attention) is an end-to-end trainable, scalable latent-memory framework that lets LLMs handle context windows up to 100M tokens, replacing RAG or fixed-size RNN-style memory with a differentiable sparse-attention memory bank.

  26. 29

    ZDTaichu5.0-9B

    3.1KTaichu-AI/ZDTaichu5.0-9B · Python

    ZDTaichu5.0-9B is a multimodal foundation model that combines a Qwen3.5-9B language backbone with a C-RADIOv4-H vision encoder for general visual understanding, spatial reasoning, embodied-AI research and agentic tool use.

  27. 30

    ai-agents-the-definitive-guide

    2.8KNicolepcx/ai-agents-the-definitive-guide · Jupyter Notebook

    It is the companion code repository for the book 'AI Agents - The Definitive Guide': a set of runnable Jupyter notebooks that walk through every chapter, from basic agent patterns to production deployment and threat modeling.

  28. 31

    AnyJev

    1.4Knokia-applied-research/AnyJev · Python

    A Python library that turns any open LLM into a Jev-style decision model: ask a typed question, get back a choice with a probability you can threshold — no generation, no parsing, no fine-tuning.

  29. 32

    open-ontologies

    1.1Kfabio-rovai/open-ontologies · Rust

    Open Ontologies lets you plan, apply and roll back changes to a production ontology: it shows every consequence as a blast-radius report before you apply a change, then hands the reviewer a proof they can re-check without trusting you or the engine.

  30. 33

    GLM-5.3-Flash-J-Space-Capability-Realization-Report

    1KTiger3807861189/GLM-5.3-Flash-J-Space-Capability-Realization-Report

    J-Space Cognition Suite is a model-agnostic inference-time control system for deep reasoning, long-horizon work, tool use, verification, and recovery, packaged as a Skill that loads into AI agent hosts.

  31. 34

    TypeLLM

    956TypeLLM/TypeLLM · Python

    TypeLLM adds type-safe, schema-guaranteed output generation to existing open autoregressive LLMs, without changing their weights or architecture, letting models still think freely while returning typed JSON Schema-constrained answers.

  32. 35

    awesome-jev

    601kydlikebtc/awesome-jev · Python

    A curated index of 1207 public resources for Jev, TypeSafe AI's System One decision model, organised by the decision each resource illustrates. It's an independent catalog, not the product or an SDK.

  33. 36

    StartLux-Decision

    552StartLuxLabs/StartLux-Decision · Python

    StartLux-Decision is a family of typed decision models (five dense sizes from 0.8B to 27B plus a 35B-A3B mixture-of-experts) that answer a question with a probability for each option instead of generating text.

  34. 37

    Ai-jailbreak

    476Quincunx33/Ai-jailbreak · JavaScript

    A curated collection of jailbreak prompts and exploit techniques that bypass the safety guardrails of both open-weight models (Gemma, Qwen, Llama, DeepSeek, Phi-4, Mistral) and frontier closed models (GPT-5.x, Gemini 3.x, Grok 4.x), documented for red-teaming and AI safety research.

  35. 38

    jeeves

    425PostHog/jeeves · Python

    Jeeves is a reasoning-enhanced 'Jev-like' decision model built on Qwen3.5-9B with LoRA. It thinks through a reasoning chain before giving calibrated yes/no, multiple-choice, or rating answers, and ships with a diffusion drafter for faster inference.

  36. 39

    Hy4-preview

    395Tencent-Hunyuan/Hy4-preview · Python

    Hy4 preview is Tencent Hunyuan's newest open-source large language model, a 770B-parameter Mixture-of-Experts model (49B active per token) built for coding, office/data analysis, game development, and scientific research tasks.

  37. 40

    dsh-commandcode-provider

    393Mars-Sea/dsh-commandcode-provider · TypeScript

    This is an unofficial plugin that adds Command Code as an LLM provider inside DeepSeek Harness (dsh), giving dsh access to Command Code's live model catalog, plan-aware model selection, reasoning effort control, image input, web search, and multi-account support.

  38. 41

    EvoVLM

    319zzzz7788990213-ops/EvoVLM · Python

    EvoVLM is a research project that runs LLM-guided, multi-objective evolutionary search over vision-language inference programs. It targets the accuracy-latency trade-off on scientific chart understanding while keeping model weights frozen.

  39. 42

    Astra-Ares

    308miuuyy/Astra-Ares · JavaScript

    Astra-Ares is an experimental tool that lets Jev, an AI evaluator, adapt GPT-6's reasoning effort while a Codex task runs, matching thinking depth to each next step to reduce token usage.

  40. 43

    imp

    254deepfates/imp · Elixir

    Imp is a full port of DSPy to the BEAM: an Elixir framework for building and automatically improving language-model programs. You declare a step's inputs and outputs as a typed signature, and an optimizer tunes its prompts or examples against labeled data.

  41. 44

    dsh-reasoning-effort

    227HanaAyane/dsh-reasoning-effort · TypeScript

    It is a plugin for DeepSeek Harness (DSH) that brings a Codex-style model-and-reasoning-effort slider into the chat input area, complete with an animated 8-frame running 'fat fish' icon that speeds up as you drag.

  42. 45

    arc-code

    143jerber/arc-code · Python

    arc-code is a research harness that lets Claude Code (or Codex) play the ARC-AGI-3 benchmark games through a shell and filesystem, writing its own parsers, simulators and search code for each game instead of relying on built-in ARC-specific tooling.

  43. 46

    Iris

    102AllSpark-Research/Iris · Python

    Iris is a pair of open-weight search agent models (Iris-mini, 35B-A3B, and Iris-pro, 397B-A17B) post-trained from the Qwen3.5/3.6 series to decide what to search, how to read results, when to keep searching, and when enough evidence has been gathered.

Often tagged together