← back to the directory
Model Inference & Serving

Yi

1Bilingual 2open-weight 3LLM 4series 5from 601.AI

the six Ws · specification

W1 Who

01.AI (the startup founded by Kai-Fu Lee) developed the Yi model series.

W2 What

Yi is a family of bilingual English/Chinese large language models, including Yi-34B and Yi-1.5, released under Apache 2.0.

W3 Where

Weights are hosted on Hugging Face and ModelScope, with code on GitHub at 01-ai/Yi.

W4 When

Yi-34B released in November 2023, with Yi-1.5 following in May 2024.

W5 Why

Built to provide strong bilingual reasoning and long-context (200K token) performance as an open alternative to closed LLMs.

W6 With

Runs via Hugging Face Transformers, vLLM, or llama.cpp-compatible GGUF conversions.

W7 Watch

Apache 2.0 license, weights and code fully open with no gating, no credentials required, runs locally or via third-party API providers; maintained by 01.AI.

alternatives

works with

01.AIOpen-weightsLLMApache-2.0BilingualLong-context

for agents & scripts

Reading this as a machine? Query it directly.

Search is open JSON - no key. Report telemetry after using a tool and it feeds that tool’s Proof Score. Or speak MCP to /mcp and discover tools mid-loop.