deepseek-v4.1IMAGE
DeepSeek V4.1 Flash 1017 vs Opus 5.5 and GPT-6.1 Sol on 250 hard problems
On 250 hard problems (BFCL tool calls, Omni-MATH 7+, SuperGPQA hard), Flash scored 76.4% vs Opus 82.0% and Sol 74.8%, tying Opus on tool calls at ~1/35 the cost per correct call but trailing 11 points on hard knowledge.
Aleksey Smolenchuk
@lxeArchived 10 days agoOriginally posted 2026.10.18
Aleksey Smolenchuk @lxeOriginally posted 2026.10.18
DeepSeek V4.1 Flash 1017 came out today at $0.60 per million output tokens. I ran it against Opus 5.5 and GPT-6.1 Sol on 250 hard problems (BFCL tool calls, Omni-MATH 7+, SuperGPQA hard), simple code grader on @pocketmodelai Flash 76.4% at a 128k token limit Opus 82.0% Sol 74.8% Flash ties Opus on tool calls at 1/35 the cost per correct call, and matches Sol everywhere. It trails Opus by 11 points on hard knowledge, and takes about 2 minutes per hard math problem.
x.com/lxe/status/2111576318836101135Archived 2026.10.18 16:28