winfunc/deepreasoning

Health v2 · maintenance only

Adoption not part of the score

5363 stars · 437 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

DeepReasoning is a high-performance Rust-based LLM inference API and chat UI that combines DeepSeek R1's chain-of-thought reasoning traces with Anthropic Claude's creative and code generation capabilities in a single unified stream. It supports self-hosting and BYOK (bring your own key) usage, keeping API keys and data under user control.

Use cases

• combine deepseek r1 reasoning with claude responses in one api
• self-host an llm chat interface with my own api keys
• get chain-of-thought reasoning traces followed by a claude answer
• stream reasoning and final answers from two llm providers together
• run a private chatgpt-style ui without sending data to third parties
• improve claude's math and logic answers with r1-style reasoning

When to choose

• you want R1's reasoning quality combined with Claude's coding and conversational strengths
• you need a self-hosted, privacy-focused inference proxy with local key management
• you want a single streaming API that merges two model providers' outputs
• you prefer a high-performance Rust backend with a ready-made chat UI

When to avoid

• you need fully local/offline inference since it depends on external DeepSeek and Anthropic APIs
• you want a single-provider solution without multi-model orchestration
• you require official vendor support or affiliation with Anthropic or DeepSeek
• you need fine-tuning or model training capabilities

Facets

service · maturity active

llm-inference http-server api-framework chatbot rag large-language-models artificial-intelligence developer-tools rust self-hosted cross-platform deepseek-r1 anthropic-claude chain-of-thought reasoning byok chat-ui streaming ai-agents web-server

Member repositories

For agents

markdown · JSON · MCP: product_card(name="winfunc/deepreasoning")