$0.48 and 43 minutes: DeepSeek V4.1 builds an interactive map, igniting cost vs capability debate
teortaxesTex · x · 2026-10-25
- A user (@tphuang) generated an interactive map of China's Middle Corridor bypassing Turkey via the Caucasus and Bulgaria to Budapest with a single DeepSeek prompt — 43 minutes, $0.48 total, concluding "the limitation now is fully on me, not on DeepSeek."
- Screenshots show OpenCode's cost/session leaderboard: GLM-5.2 tops it at $2.32/session while DeepSeek V4-Pro sits at $0.667; tphuang argues DeepSeek spent two years optimizing cost while Z.ai focused on top-tier SWE experience users happily pay for.
- teortaxesTex pushes back on the "DS only cares about cheap" take: until V4 the API was unusually slow because the real goal was scaling RL at very long contexts. He cites the DeepSeek V4 paper — 1.6T params (49B activated), 1M-token context, CSA+HCA hybrid attention, MHC residual connections, Muon optimizer.
- He adds that Liang's statements should be taken literally: DeepSeek may have deliberately spent 1.5 years building a scalable platform before shipping a "preview" LLM.
More from Infra
- Amazon's ATTUNE coding agent lifts LLM training throughput 1.7x on H100/H200/B200 clusters — kalyan_kpl · 2026-10-25
- HKUST & Huawei's BalanceRoute tackles load balancing in MoE LLM serving — jiqizhixin · 2026-10-25
- Pushing local AI video to the limit: MiniMax H3 running on a single RTX 3090 — DeliciousGorilla · 2026-10-25
- a16z: GPU rental rates defy depreciation, H200 up 143% YTD — elonmusk · 2026-10-25
- Qwen3.8 Flash Next IQ3_S runs locally on 16GB VRAM at 250 tok/s generation, 5,600 tok/s prefill — yehyakar · 2026-10-25
- 4x DGX Spark cluster recipe runs GLM-4.6-Flash with 1M context, 16 concurrent, fully local — AIFlow_ML · 2026-10-25