deepseek-chat
deepseek-chat now serves DeepSeek-V3 with the same API surface.
DeepSeek's change log says: The deepseek-chat model has been upgraded to DeepSeek-V3. The API remains unchanged. You can invoke DeepSeek-V3 by specifying model='deepseek-chat'. For details, please refer to: introducing DeepSeek-V3
Who it affects: API users calling model=deepseek-chat.
What it means: deepseek-chat now serves DeepSeek-V3 with the same API surface.
Official sources
Related pages: /guides/model-name-migration · /models/deepseek-v3 · /news
Context for developers
The 2024-12-26 update upgraded deepseek-chat to DeepSeek-V3 while keeping request/response fields compatible. Developers who hard-coded the alias received V3 capabilities (MoE architecture, MLA, FP8 training stack described in the V3 README) without changing client code.
DeepSeek's later 2026 migration guides replace chat/reasoner aliases with explicit V4 model IDs. Treat this article as a milestone in that timeline; new projects should call deepseek-flash or deepseek-v4-pro per the current pricing table.
Independent resource, not affiliated with DeepSeek.