DeepSeek V4.1 Flash hands-on tests and comparisons

Claims DeepSeek V4.1-Flash surpasses Claude Opus 5 on coding at about $0.2 per task versus over $10 for Claude, releasing 763 billion parameters for free under the MIT license.

The post says the release improves 3D, coding, agents and vision, ships faster and shorter outputs at the same price with 1M context, is live on app, web and API, and has MIT-licensed open weights.

Official figures show Terminal-Bench 4.0 at 42.3 vs 31.2 and DeepSWE v1.1 at 77.0, beating Opus 5's 74.0; the API now serves deepseek-flash.

DeepSWE 77% vs Opus 5's 74%, demos include a Forbidden City 3D scene and a Godot Portal 2 remake; Search/Fetch API bills 5 yuan per 1,000 successful calls, but users say it follows instructions literally and won't add UI/UX polish.

A user reports DeepSeek V4.1 Flash 1017 scoring 77% on DeepSWE at a much lower price than closed models, citing open weights and agentic gains.

Official release: 763B params, 1M context, MIT open weights; DeepSWE 77.0 vs Opus 5's 74.0 at about $0.2 per task versus $10+, though it still trails on HLE and Terminal-Bench.

A rundown of the official release: MIT-licensed weights, native image-text FP8, 1M context and 256K output, DeepSWE 77 beating Opus 5 while trailing on Terminal-Bench and HLE, plus Search & Fetch at $0.75 per 1,000 calls and free in Harness, Claude Code and Codex until November 1.

The author argues self-feedback reinforcement learning lets V4.1 Flash 1017 build complete projects from scratch in game dev and 3D modeling, enabling wish-style coding.

The deepseek-flash call name and pricing are unchanged so no code changes are needed; Terminal-Bench 4.0 rose from 31.2 to 42.3, coding gains are largest, and outputs are shorter, saving tokens.

A user says the 1017 update is stronger and cheaper, producing games and 3D work (Forbidden City 3D, a racing game, a Portal 2 remake) from scratch, plus a search API at 5 yuan per thousand calls and weights on HuggingFace.

Commentary on the surprise Saturday launch: doubts that it beats Opus 5.5, questions why it is compared to Sol 5.6 or Opus 5 when the prior V4.1 Flash scored higher, and asks when t3code will add DeepSeek Harness.

The poster points a local DeepSeek V4.1 running on two Sparks at an app holding personal decisions and finances, says it surfaced a blind spot and eased stress, and just learned of the 1017 upgrade.