WHAT NEW IN X11.01 SUN

AI collections from X · 22 collections · 63,170 posts

Back to No.020 DeepSeek V4.1 Flash hands-on tests and comparisonsUpdated 15 hours ago
deepseek-v4.1IMAGE

Reflection Beam 1.0 vs DeepSeek V4.1 Flash on 90 hard tasks: 67 vs 66

On 90 hard BFCL tool-call, Omni-MATH 7+ and SuperGPQA items graded by code, Beam scored 67/90 and DeepSeek 66/90, though Beam costs about 6x more.

Aleksey Smolenchuk
@lxeArchived 11 days agoOriginally posted 2026.10.20
Reflection Beam 1.0 vs DeepSeek V4.1 Flash on 90 hard tasks: 67 vs 66
Aleksey Smolenchuk @lxeOriginally posted 2026.10.20
Open original on X

Reflection @reflection_ai Beam 1.0 vs DeepSeek V4.1 Flash 1017 on 90 hard specimens (BFCL tool calls, Omni-MATH 7+, SuperGPQA hard), one sample each, code grader on @pocketmodelai: Beam 67/90
DeepSeek 66/90 Beam is 6x the cost. Maybe a lossy quant similar to my dsx experiment can help equalize the pricing here. They are similarly sized models.

x.com/lxe/status/2112317533860839504Archived 2026.10.20 12:05
Open original on X