Search Results for "reasoning" - Page 10
Sort By:
Showing 271 open source projects for "reasoning"
View related business solutions-
MongoDB Atlas runs apps anywhereMongoDB Atlas gives you the freedom to build and run modern applications anywhere—across AWS, Azure, and Google Cloud. With global availability in over 115 regions, Atlas lets you deploy close to your users, meet compliance needs, and scale with confidence across any geography.
-
Demo Series - Small Business Backup By VeeamWatch this on-demand demo series and learn how to protect your Microsoft 365 data with clear, simple, actionable steps that are easy to implement for businesses of all sizes.
-
1
DeepSeek VL
Towards Real-World Vision-Language Understanding
DeepSeek-VL is DeepSeek’s initial vision-language model that anchors their multimodal stack. It enables understanding and generation across visual and textual modalities—meaning it can process an image + a prompt, answer questions about images, caption, classify, or reason about visuals in context. The model is likely used internally as the visual encoder backbone for agent use cases, to ground perception in downstream tasks (e.g. answering questions about a screenshot). The repository... -
2
ToRA
Tool-integrated Reasoning LLM Agents
ToRA is an open-source framework developed by Microsoft for building tool-integrated reasoning agents powered by large language models. The project focuses on improving the ability of AI systems to solve complex mathematical and analytical problems by combining natural language reasoning with external computational tools. Instead of relying solely on text generation, the system dynamically invokes tools such as symbolic solvers or programming libraries when deeper computation is required. ... -
3
Prompt Engineering Interactive Tutorial
Anthropic's Interactive Prompt Engineering Tutorial
...The course leans heavily on realistic failure modes (ambiguity, hallucination, brittle instructions) and shows how to iteratively debug prompts the way you would debug code. Lessons include building prompts from scratch for common tasks like extraction, classification, transformation, and step-by-step reasoning, with checkpoints that let you compare your outputs against solid baselines. You’ll also practice advanced patterns such as tool use, constrained generation, and response validation so outputs are trustworthy and machine-consumable. -
4
Evals
Evals is a framework for evaluating LLMs and LLM systems
...It includes utilities and APIs to plug in completion functions, manage prompts, wrap retries or error handling, and register new evaluation types. It also maintains a growing registry of standard benchmarks or “evals” that users can reuse (for example, tasks measuring reasoning, factual accuracy, or chain-of-thought capabilities). The design is modular so you can extend or compose new evals, integrate with your own model APIs, and capture rich metadata about each run (prompt, responses, metrics). -
Build Agents and Models on One PlatformGemini Enterprise Agent Platform is Google Cloud's comprehensive platform for developers to build, scale, govern, and optimize agents and models. Choose from Google's most advanced models and third-party models like Anthropic's Claude Model Family.
-
5
Qwen-VL
Chat & pretrained large vision language model
...It accepts image inputs (with optional bounding boxes) and text, and produces text (and sometimes bounding boxes) as output. The model variants (VL-Plus, VL-Max, etc.) have been upgraded for better visual reasoning, text recognition from images, fine-grained understanding, and support for high image resolutions / extreme aspect ratios. Qwen-VL supports multilingual inputs and conversation (e.g. Chinese, English), and is aimed at tasks like image captioning, question answering on images (VQA, DocVQA), grounding (detecting objects or regions from textual queries), etc. -
6
RAG-Retrieval
Unify Efficient Fine-tuning of RAG Retrieval, including Embedding
RAG-Retrieval is an open-source framework for building and training retrieval systems used in retrieval-augmented generation pipelines. Retrieval-augmented generation combines large language models with external knowledge retrieval to improve factual accuracy and domain-specific reasoning. This repository provides end-to-end infrastructure for training retrieval models, performing inference, and distilling embedding models for improved performance. It includes implementations of modern embedding architectures designed to map documents and queries into vector spaces for efficient similarity search. The framework also supports reranking models that refine retrieved results using large language models or lightweight transformer architectures. ... -
7
Tensor Puzzles
Solve puzzles. Improve your pytorch
...Standard convenience functions such as sum, view, squeeze, and take are intentionally prohibited. This constraint encourages a deeper understanding of shapes, indexing, and vectorized computation. Automated tests and visual examples help learners check their reasoning as they progress. The notebook is designed for Google Colab and is accompanied by a video walkthrough. -
8
Alpaca-CoT
We unified the interfaces of instruction-tuning data
Alpaca-CoT is an open research project focused on improving reasoning capabilities in language models through chain-of-thought training data. The project builds upon the Alpaca instruction-tuning approach by introducing datasets and methods that encourage models to produce intermediate reasoning steps when solving problems. Instead of generating answers directly, the model learns to produce logical reasoning sequences that lead to the final solution. -
9
MGIE
Guiding Instruction-based Image Editing via Multimodal Large Language
MGIE—Guiding Instruction-based Image Editing—demonstrates how a multimodal LLM can parse natural-language editing instructions and then drive image transformations accordingly. The project focuses on making edits explainable and controllable: the model interprets text guidance, reasons over image content, and outputs edits aligned with user intent. It’s positioned as an ICLR 2024 Spotlight work, with code and references that show how to connect language planning to concrete image operations.... -
$300 Free Credits to Build on Google CloudPut your $300 in credit toward real workloads, then keep building with free monthly usage for 20+ products. No commitment and no charge until you upgrade.
-
10
Ailice
AIlice is a fully autonomous, general-purpose AI agent
...AIlice is designed for a wide range of workloads, including coding, thematic research, literature analysis, system management, and mixed workflows that require several reasoning modes at once. -
11
InternLM
Official release of InternLM series
...Beyond model weights, the project emphasizes an ecosystem view, pointing developers to compatible tools and projects across training and inference so teams can build end-to-end workflows. InternLM’s direction includes strong general-purpose capabilities and ongoing iterations that target improved reasoning, coding, and tool-use behaviors. The broader InternLM ecosystem also includes training tooling and guidance aimed at making fine-tuning and adaptation more accessible across hardware setups, including smaller single-GPU environments and larger multi-node configurations. -
12
BIG-bench
Beyond the Imitation Game collaborative benchmark for measuring
BIG-bench (Beyond the Imitation Game Benchmark) is a large, collaborative benchmark suite designed to probe the capabilities and limitations of large language models across hundreds of diverse tasks. Rather than focusing on a single metric or domain, it aggregates many hand-authored tasks that test reasoning, commonsense, math, linguistics, ethics, and creativity. Tasks are intentionally heterogeneous: some are multiple-choice with exact scoring, others are free-form generation judged by model-based or human evaluation. The suite provides a common JSON task format and an evaluation harness so research groups can contribute new tasks and reproduce results consistently. ... -
13
Graph of Thoughts
Official Implementation of "Graph of Thoughts
Graph of Thoughts is an open-source framework that implements a novel reasoning paradigm for large language models by organizing reasoning steps as a structured graph instead of a simple linear chain. Traditional reasoning methods such as chain-of-thought generate sequential reasoning steps, but Graph of Thoughts introduces a more flexible structure where multiple reasoning paths can be explored and evaluated simultaneously. -
14
CoTracker
CoTracker is a model for tracking any point (pixel) on a video
CoTracker is a learning-based point tracking system that jointly follows many user-specified points across a video, rather than tracking each point independently. By reasoning about all tracks together, it can maintain temporal consistency, handle mutual occlusions, and reduce identity swaps when trajectories cross. The model takes sparse point queries on one frame and predicts their sub-pixel locations and a visibility score for every subsequent frame, producing long, coherent trajectories. Its transformer-style architecture aggregates information both along time and across points, allowing it to recover tracks even after brief disappearances. ... -
15
Nougat
Implementation of Nougat Neural Optical Understanding
...Because it integrates structured layout reasoning, Nougat tends to produce more compositional and controllable results than purely unconstrained generative models. -
16
ReAct Prompting
Synergizing Reasoning and Acting in Language Models
ReAct is an open-source research project that demonstrates a prompting and reasoning framework designed to improve the problem-solving capabilities of large language models. The project implements the methodology described in the research paper “ReAct: Synergizing Reasoning and Acting in Language Models,” which combines reasoning traces with action-based interactions. Instead of generating answers in a single step, models using the ReAct approach produce intermediate reasoning steps and perform actions such as searching for information or interacting with external tools. ... -
17
ThoughtSource
A central, open resource for data and tools
ThoughtSource is a central, open resource and community centered on data and tools for chain-of-thought reasoning in large language models (Wei 2022). Our long-term goal is to enable trustworthy and robust reasoning in advanced AI systems for driving scientific research and medical practice. -
18
Repo of Tree of Thoughts (ToT)
Implementation of "Tree of Thoughts
...To surmount these challenges, we introduce a new framework for language model inference, Tree of Thoughts (ToT), which generalizes over the popular Chain of Thought approach to prompting language models and enables exploration over coherent units of text (thoughts) that serve as intermediate steps toward problem-solving. ToT allows LMs to perform deliberate decision-making by considering multiple different reasoning paths and self-evaluating choices to decide the next course of action, as well as looking ahead or backtracking when necessary to make global choices. -
19
CausalNex
A Python library that helps data scientists to infer causation
CausalNex is a Python library that uses Bayesian Networks to combine machine learning and domain expertise for causal reasoning. You can use CausalNex to uncover structural relationships in your data, learn complex distributions, and observe the effect of potential interventions. -
20
book
A textbook on informal homotopy type theory
The canonical Homotopy Type Theory informal textbook, initiated at the Institute for Advanced Study (2012–13) under the Univalent Foundations project. It presents the theory pedagogically without requiring technical proof assistant use. -
21
PRM800K
800,000 step-level correctness labels on LLM solutions to MATH problem
...The repository releases the raw labels and the labeler instructions used in two project phases, enabling researchers to study how human raters graded intermediate reasoning. Data are stored as newline-delimited JSONL files tracked with Git LFS, where each line is a full solution sample that can contain many step-level labels and rich metadata such as labeler UUIDs, timestamps, generation identifiers, and quality-control flags. Each labeled step can include multiple candidate completions with ratings of -1, 0, or +1, optional human-written corrections (phase 1), and a chosen completion index, along with a final finish reason such as found_error, solution, bad_problem, or give_up. -
22
Chameleon LLM
Codes for "Chameleon: Plug-and-Play Compositional Reasoning
Discover Chameleon, our cutting-edge compositional reasoning framework designed to enhance large language models (LLMs) and overcome their inherent limitations, such as outdated information and lack of precise reasoning. By integrating various tools such as vision models, web search engines, Python functions, and rule-based modules, Chameleon delivers more accurate, up-to-date, and precise responses, making it a game-changer in the natural language processing landscape. ... -
23
CodeContests
Large dataset of coding contests designed for AI and ML model training
CodeContests, developed by Google DeepMind, is a large-scale competitive programming dataset designed for training and evaluating machine learning models on code generation and problem solving. This dataset played a central role in the development of AlphaCode, DeepMind’s model for solving programming problems at a human-competitive level, as published in Science. CodeContests aggregates problems and human-written solutions from multiple programming competition platforms, including AtCoder,... -
24
Grade School Math
8.5K high quality grade school math problems
The grade-school-math repository (sometimes called GSM8K) is a curated dataset of 8,500 high-quality grade school math word problems intended for evaluating mathematical reasoning capabilities of language models. It is structured into 7,500 training problems and 1,000 test problems. These aren’t trivial exercises — many require multi-step reasoning, combining arithmetic operations, and handling intermediate steps (e.g. “If she sold half as many in May… how many in total?”). The problems are written by human authors (not automatically generated) to ensure linguistic variety and realism. ... -
25
Differentiable Neural Computer
A TensorFlow implementation of the Differentiable Neural Computer
The Differentiable Neural Computer (DNC), developed by Google DeepMind, is a neural network architecture augmented with dynamic external memory, enabling it to learn algorithms and solve complex reasoning tasks. Published in Nature in 2016 under the paper “Hybrid computing using a neural network with dynamic external memory,” the DNC combines the pattern recognition power of neural networks with a memory module that can be written to and read from in a differentiable way. This allows the model to learn how to store and retrieve information across long time horizons, much like a traditional computer. ...