DeepSeek V4.1 Flash 1017 vs Qwen 3.8 Flash Next
aka DeepSeek 1017 vs Qwen Next
- Author picksDeepSeek V4.1 Flash 10171–0
- FasterQwen 3.8 Flash Next1/1 postsmedian gap ~1.1×
As stated in each post, not a controlled benchmark.
Posts
4
Voxel world on one Mac Studio, local Flash models: Qwen 3.8 Flash Next, DeepSeek V4.1 Flash 1017. Posted by @niklaslenz_ai.
Voxel reef, same prompt, two models in DeepSeek Harness: Qwen 3.8 Flash Next, DeepSeek V4.1 Flash 1017. Posted by @zbian73.
3D Orbital Garden, three models ranked: Qwen 3.8 Flash Next, DeepSeek V4.1 Flash 1017, Claude Opus 5.5. Posted by @ZenMagnets.
Same prompt, separate runs
1Not a controlled comparison: prompts, tools and time differ.
About this matchup
DeepSeek V4.1 Flash 1017 vs Qwen 3.8 Flash Next in real side-by-side tests: 4 posts from X where both got the same prompt; 1 post reports cost or time. Judge coding ability from the actual outputs.
DeepSeek V4.1 Flash 1017 vs Qwen 3.8 Flash Next: which is better for coding?
Which is better for coding, DeepSeek V4.1 Flash 1017 or Qwen 3.8 Flash Next?
Author picks: where the author preferred one of the two, they picked DeepSeek V4.1 Flash 1017 once and Qwen 3.8 Flash Next never. That covers 1 of the 4 posts here; each post is a different task, so watch the outputs above for the work you care about.
Which is faster, DeepSeek V4.1 Flash 1017 or Qwen 3.8 Flash Next?
1 post states a time for both: Qwen 3.8 Flash Next was faster in it (~1.1× faster). Times are as stated in each post; tools and settings differ between posts.
What have DeepSeek V4.1 Flash 1017 and Qwen 3.8 Flash Next been tested on?
The most-viewed of the 4 posts: voxel reef, same prompt, two models in DeepSeek Harness (@zbian73), voxel world on one Mac Studio, local Flash models (@niklaslenz_ai), 3D Orbital Garden, three models ranked (@ZenMagnets). Categories: 3D & Shaders.
Summary
DeepSeek V4.1 Flash 1017 and Qwen 3.8 Flash Next appear together in 4 posts on X: voxel Eiffel Tower on a local RTX 3090 (@nawariokra), voxel world on one Mac Studio, local Flash models (@niklaslenz_ai), voxel reef, same prompt, two models in DeepSeek Harness (@zbian73) and 3D Orbital Garden, three models ranked (@ZenMagnets). 1 of them are part of larger lineups with other models. In @niklaslenz_ai's voxel world on one Mac Studio, local Flash models, Qwen 3.8 Flash Next was faster (58.1 min vs 1 h 06 min) and DeepSeek V4.1 Flash 1017 used fewer tokens (168.2K tokens vs 225.9K tokens), as reported in the post.
What each post reports
Only numbers stated in each post. “—” means the post doesn’t say.
Quick facts
| Model | Vendor | Released | Input $/1M tokens | Output $/1M tokens | Context (tokens) | Official links |
|---|---|---|---|---|---|---|
| DeepSeek V4.1 Flash 1017 | DeepSeek | 2026-10-17 | $0.30peak hours; off-peak is half price | $1.20peak hours; off-peak is half price | 1,000,000 | Release notes · Pricing |
| Qwen 3.8 Flash Next | Alibaba (Qwen) |
From the vendors’ official pages, checked 2026-10-19. Standard API list prices; each value links its source and a blank cell could not be verified. The archive’s own numbers above are what each post states.
Who won where
2- Voxel world on one Mac Studio, local Flash modelsThe author likes the DeepSeek world much better, even though Qwen decodes about 2.4 times faster.— the author’s view, @niklaslenz_ai (paraphrased)
- 3D Orbital Garden, three models rankedOpus 5.5 ranks first, DeepSeek V4.1 Flash 1017 second (the best 3D reasoning the author has seen from an open model), Qwen third.— the author’s view, @ZenMagnets (paraphrased)
Models
Related matchups
Claude Opus 5.5 vs DeepSeek V4.1 Flash 1017123 posts · 519.4K viewsDeepSeek V4.1 Flash 1017 vs MiniMax M3.1 Flash91 posts · 115.3K viewsDeepSeek V4.1 Flash 1017 vs Gemini 4 Argon40 posts · 307.4K viewsDeepSeek V4.1 Flash 1017 vs GPT-6.1 Sol35 posts · 488.8K viewsDeepSeek V4.1 Flash 1017 vs GPT-6.1 Terra31 posts · 69.9K viewsDeepSeek V4.1 Flash 1017 vs GPT-6 Astra22 posts · 112.9K views