WHAT NEW IN X10.30 FRI

AI collections from X · 22 collections · 61,986 posts

deepseek-v4.1VIDEO

DeepSeek V4.1 Flash vs Opus 5.5 on a pagoda build test: cost and output compared

On the same Agents.md task, DeepSeek in DSH took ~35 min, 42M tokens and ~$0.33 versus Opus 5.5 in CC at ~25 min, 15.5M tokens and ~$0.25, yet the author preferred DeepSeek's result.

Leo · Building w/ AI
@leo114119Archived 12 days agoOriginally posted 2026.10.18
DeepSeek V4.1 Flash vs Opus 5.5 on a pagoda build test: cost and output compared (image 1 of 2)
DeepSeek V4.1 Flash vs Opus 5.5 on a pagoda build test: cost and output compared (image 2 of 2)
Leo · Building w/ AI @leo114119Originally posted 2026.10.18
Open original on X

Deepseek v4.1 Flash 1017 与 Opus 5.5 的宝塔测试 A) DeepSeek 在 DSH 中的 Agents.md。≈35 分钟,4200 万令牌,≈$0.33 B) Opus 5.5 在 CC 中的 Agents.md。≈25 分钟,仅 1550 万令牌,≈$0.25。但更简单 出乎意料的,DeepSeek居然花了更多的成本。。应该是多轮思考导致的额外支出。。。 这就是俗话说的:勤能补拙? opus 5.5 毫无疑问更强大,但这两个视频,我居然觉得 ds的好像更好一点。。。 https://x.com/teortaxesTex/status/2111641292761714741/video/1 https://x.com/teortaxesTex/status/2111641292761714741/video/2

x.com/leo114119/status/2111646447406064017Archived 2026.10.18 16:26