deepseek-v4.1IMAGE
DeepSeek V4.1 Flash 1017 vs Opus 5: 77.0 vs 74.0 on DeepSWE, loses Terminal-Bench
The post cites 1M context, image input and MIT licensing, plus numbers showing DeepSWE 77.0 vs 74.0 and Automation-Bench 60.5 vs 50.3 ahead of Opus 5, while Opus 5 wins Terminal-Bench 4.0.
painn
@painn_xArchived 10 days agoOriginally posted 2026.10.17
painn @painn_xOriginally posted 2026.10.17
DeepSeek V4.1 Flash 1017 beats Opus 5 and its FREE 😳 ngl, DeepSeek might actually be cooking with this one what you get: - 1M context window - image input support - MIT license, fully open source their numbers vs Opus 5: > DeepSWE: 77.0 vs 74.0 > Automation-Bench: 60.5 vs 50.3 > Terminal-Bench 4.0: Opus 5 wins the whole thing costs nothing. no card. use it free: http://chat.deepseek.com open weights: https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017
x.com/painn_x/status/2111458608340718029Archived 2026.10.18 16:36