Deepseek Chat

released 2025-12-01

BenchAtlas Index

as of 2026-09-11

Not enough benchmark coverage for a composite score (needs ≥4 families across ≥3 categories). The per-benchmark evidence below stands on its own — sparse coverage is an evidence gap, not a low score.

Benchmark evidence

3 results

external indices

ECI

146.2 points

independent

Epoch AI Benchmarking Hub

knowledge science

GPQA Diamond

71.2 %

independent

Epoch AI Benchmarking Hub

2026-07-16

reasoning math

OTIS Mock AIME 2024–2025

48.9 %

independent

Epoch AI Benchmarking Hub

2026-07-16

Agent + model results

systems, not bare-model scores

agent + model Moatless Tools + Deepseek Chat

SWE-bench Lite

30.7 %

community

SWE-bench Leaderboard

These scores measure the whole agent system (scaffold, tools, budgets) — they are never merged into the bare model’s numbers.