← Model explorer

Sourced model profile · Current through July 14, 2026

DeepSeekMath-V2

An unrestricted open-weight DeepSeek model documented for mathematical reasoning, theorem proving, and proof generation and verification.

Developed byDeepSeek
Released
November 27, 2025

What the evidence says

These facts come from the reviewed Epoch AI catalog artifact and its linked primary record. They describe the model. They do not rank it.

Developer [1]
DeepSeek
Released [1]
November 27, 2025
Domains [1]
Language
Tasks [1]
Mathematical reasoning · Theorem proving · Proof generation · Proof verification
Access [1]
Open weights (unrestricted)
Weights [1]
Open

Where it may fit

These are evidence-bounded screening suggestions, not performance claims or purchase recommendations.

Formal mathematics research

Consider it when theorem proving or proof verification is the explicit task and open weights are required for inspection.

Evidence fields: tasks, openWeights [1]

Proof-workflow tooling

Consider it for building research tooling around proof generation with weights the team can host directly.

Evidence fields: tasks, access [1]

Limitations

The record documents task scope only. It does not report proof accuracy, benchmark results, hardware needs, context window, latency, or safety evaluations, and no exact-match Arena rating exists for it.

Current standing

According to the LMArena snapshot published July 12, 2026, no exact match for this catalog record appears in the style-controlled text category, so this page reports no current rating for DeepSeekMath-V2.