DeepSeek · DeepSeek V4
Short-lived experimental image-input version of V4 Flash; its API id now routes to V4.1 Flash. Listed only for history.
Leader: Claude Opus 5.5 at 58
Leader: Claude Sonnet 5.5 at 70.6%
Leader: Claude Fable 5.1 at 91.4%
Leader: Claude Opus 5.5 at 66.9%
Leader: Muse Wonder Full at 1857
Leader: GPT-6 Astra at 96.1%
Leader: Claude Sonnet 5.5 at 64.5%
Leader: Muse Wonder Full at 33%
Leader: Kimi K3 at 88.7%
Leader: Claude Opus 5.5 at 87.7%