DeepSeek V4.1 Flash hands-on tests and comparisons

The poster says the new V4.1 Flash costs just over one yuan per DeepSWE task and produced a 3D Forbidden City and a racing game from single prompts, then asks whether people would pick pricey flagships or this cheap strong Flash.

A user running V4.1 Flash with DSH on local documents reports it first searches all local files then the web for unknown concepts, and feels faster and smarter.

The user reports Terminal-Bench rising from 31.2 to 42.3, faster tasks at the same price, and clearly better one-shot 3D scenes than the old version, while noting these are official numbers pending own tests.

The post claims V4.1 Flash 1017 scores 42.8 on a terminal benchmark and reaches SOTA on DeepSWE, ahead of GPT-5.6 Sol, Kimi K3 and GLM 5.3, with low pricing, shorter outputs, 1M context and unchanged API rates.

Per DeepSeek's own chart, Flash costs about $0.20 per task and scores 77.0 on DeepSWE vs Opus 5's 74.0, though Opus still leads Terminal-Bench 51.8 to 42.3.

The post says it edges past Opus 5 on DeepSWE, ships MIT open weights, is live on OpenRouter, keeps the prior Flash price, and has a surprisingly low cache price.

DeepSeek calls it a Pareto improvement on performance and cost: DeepSWE 77.0 above Opus 5's 74.0 and GPT 5.6-Sol's 73.0, about $0.2 per task at max tier, with Harness and CodeBuddy adaptations.

DeepSeek's new Flash model is live across web, app and API with open weights, hitting 77% on DeepSWE at about 1.3 yuan per task and improving all 11 benchmarks.

A Japanese post says V4.1 Flash appears post-trained for Dsh, scores above Astra Fable, ships open weights on Hugging Face, and shows strong 3D CG demos.

Official notice says to call deepseek-flash, now upgraded to V4.1-Flash-1017 with unchanged usage; legacy names like deepseek-v4-flash still work but are served by the 1017 model at Flash pricing.

V4.1-Flash-1017 ships MIT-licensed weights with the same ~550B/8B-16B-active architecture, Terminal-Bench up 31 to 42 and DeepSWE 77 above Opus 5; the checkpoint is about 510GB.

Terminal-Bench 42.3 in the Claude Code harness, DeepSWE 77.0, 1M context and MIT weights, with $0.60/1M output off-peak plus step-by-step Claude Code setup and free Search/Fetch until Nov 1.