deepseek-v4.1POST

DeepSeek V4.1 Flash tops GPT-5.6 Sol on several benchmarks

The post cites 42.6 on HLE, 42.3 on Terminal-Bench and 77.0 on DeepSWE, saying the Flash model beats GPT-5.6 Sol on several benchmarks despite not being the full version.

Jaysen♨️

@_jayyy18_13kArchived 10 days agoOriginally posted 2026.10.17

Jaysen♨️ @_jayyy18_13kOriginally posted 2026.10.17

Open original on X

what a surprise from deepseek 👀 DS-V4.1-Flash is beating GPT-5.6 Sol on several benchmarks and it's not even the full version 42.6 on HLE, 42.3 on Terminal-Bench and 77.0 on DeepSWE that’s a serious showing for a flash model deepseek might have something special here

x.com/_jayyy18_13k/status/2111401389431914553Archived 2026.10.18 16:42