DeepSeek V4.1 Flash hands-on tests and comparisons

The post cites Artificial Analysis best-setting scores of 46.0 for Moonshot Kimi K3.1 Pro versus 32.5 for DeepSeek V4.1 Flash 1017, with Moonshot ahead.

A user notes the new AA Intelligence Index gives DeepSeek V4.1 Flash 32 points, tied with GLM 5.3 Max, versus 46 for Kimi K3.1 Pro and 51 for GPT-6.1 Terra Max.

On one task, Opus 5.5 took 30 minutes, GPT-6.1 Sol 10 minutes, Grok 4.7 41 minutes for $3.58, while DeepSeek V4.1 Flash 1017 took 50 minutes, 257K output tokens, for $0.36.

Decode throughput test: SGLang baseline 71 tps, tuned 103 tps, and TensorFold tuned 110 tps.

Artificial Analysis v5 shows open-weights models falling further behind; among released weights, DeepSeek V4.1 Flash 1017 (max) and GLM-5.3 (max) tie at 32, still below Claude Haiku 5.5.

Under TP4 ring, 256K input and thinking on, the knapcio SGLang recipe matched MiaAI's SGLang decode and was about 23% faster in prefill, while the TensorFold build lagged.

The author ranks recent open-weight Flash models by hands-on experience, placing DeepSeek V4.1 Flash 1017 first ahead of Ling-3.1-Flash and MiniMax-M3.1-Flash.

Hands-on: Opus 5.5 orchestrates while deepseek-v4.1-flash-1017 runs sub-agents in Claude Code; 24 parallel sub-agents cost roughly $9 on DeepSeek, which did 99% of the work.

After 100+ hours digging through eval tasks and running closed-source tasks across frontier models, the author finds V4.1 Flash often on par with frontier models and 13–20x cheaper even when 5–8% behind; on Cursor bench it is 13x cheaper with fully open weights down to the inference stack.

The author is cancelling GPT Pro over OpenAI halving limits on October 30, moving to Claude Max with Opus 5.5 as orchestrator and DeepSeek V4.1 Flash 1017 via Baseten at ~599 TPS for subagent work.

V4.1 Flash has led OpenRouter since Oct 25 at roughly 30% of token traffic, yet one GB300 rack could serve a day of its traffic; API is $0.15/$0.60 per 1M tokens off-peak.

OpenCode launches 2.0 and announces Operation Cheepseek phase 3, giving $80 of DeepSeek V4.1 Flash usage on a $10 monthly subscription, billed as the cheapest access anywhere.