DeepSeek V4 Pro 0813
DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes
ECI STANDING#22 of 266
SCORE155.47153.78–157.83 CI
FRONTIER LAGLaunched 6.2 months behindScores about what GPT-5.3 Codex did in February 2026.
SCORES LIKE THESE MODELSMuse Spark 1.2Meta AI · Aug 2026155.48GPT-5.2 ProOpenAI · Dec 2025155.38Claude Opus 4.6Anthropic · Feb 2026155.34GLM-5.3Z.ai (Zhipu AI) · Aug 2026155.25
Full Epoch Capabilities Index ranking ↗RELEASE DATEAug 12, 2026
WEIGHTSOpen weights
CONTEXT WINDOW1M
MAX OUTPUT384K
REASONINGYes
TOOL CALLINGYes
INPUTStext
OUTPUTStext
Specifications from Models.dev. Availability and limits may differ by provider. Benchmark scores come from Epoch AI under CC BY 4.0.