# Qwen 3.8 Flash Next

147 Hacker News commenters recommended Qwen 3.8 Flash Next and 33 criticized it, November 2025 to November 2026. It was recommended in 5 of 13 months. It meets the Hacker Approved badge rule.

**Verdict:** An open-weight model that runs fast on consumer GPUs and Macs, with coding ability some call near frontier models; it overthinks by default and degrades at low quantization.

## Why people recommend it

- Runs locally on consumer hardware (12 comments): "Setup was easier than I expected. I had it running locally in under 20 minutes, and the speed on my RTX 5070 is genuinely impressive." ([source](https://news.ycombinator.com/item?id=49963537)) "you could use Qwen Flash Next at some reasonable quant, with the new SSD hack for only keeping some of the model resident in memory" ([source](https://news.ycombinator.com/item?id=49834488)) "qwen3.8 flash next is currently running on 10gb 3080 + 64 gb ddr5 ram at respectable 40-70t/s with 1200pp at max context" ([source](https://news.ycombinator.com/item?id=50148676))
- Fast token generation (8 comments): "I've tried it but so far I have not seen anything that overly impressed me in terms of accuracy, though the speed is definitely there." ([source](https://news.ycombinator.com/item?id=49957786)) "Setup was easier than I expected. I had it running locally in under 20 minutes, and the speed on my RTX 5070 is genuinely impressive." ([source](https://news.ycombinator.com/item?id=49963537)) "qwen3.8 flash next is currently running on 10gb 3080 + 64 gb ddr5 ram at respectable 40-70t/s with 1200pp at max context" ([source](https://news.ycombinator.com/item?id=50148676))
- Strong coding, close to Opus (7 comments): "In my private C coding benchmark, Strata's Qwen Flash Next Coder IQ1_M beats Luna low and runs at 125 tok/s on my 5090" ([source](https://news.ycombinator.com/item?id=49960046)) "For example on programming, terminal tasks, troubleshooting etc Qwen3.8-Flash-Next is almost as good as opus 4.8." ([source](https://news.ycombinator.com/item?id=50219784)) "but it produces syntax correct and grammatically correct code comments, changelogs and readme type files" ([source](https://news.ycombinator.com/item?id=49494558))
- Beats smaller models for size (5 comments): "Now you can run a nerfed 125B parameter model on under $800 of hardware, and it beats a less-nerfed 27b model." ([source](https://news.ycombinator.com/item?id=49956426)) "Models like Qwen 27B 3.8 and Flash Next 3.8 show there is a lot of room to pack more into less parameters." ([source](https://news.ycombinator.com/item?id=50156696)) "This is game changing. My R9700 32GB is now smarter and about 2x faster than using Qwen-3.8-27B." ([source](https://news.ycombinator.com/item?id=49960283))

## What people warn about

- Overthinks, set effort to medium (4 comments): "Known issue with this model, I recommend setting thinking to 'medium' instead of default 'xhigh'." ([source](https://news.ycombinator.com/item?id=49961097)) "Qwen3.8 FN is deeply convinced that it's going to get extra credit for doing more than it is asked." ([source](https://news.ycombinator.com/item?id=50250081)) "Overall I'm impressed with it, but setting the /effort to medium is what you usually want" ([source](https://news.ycombinator.com/item?id=49451556))
- Degrades at low quantization (2 comments): "it generated so many more bugs, and when the quantization is lower the reasoning traces get longer" ([source](https://news.ycombinator.com/item?id=50325169)) "I'm getting mixed results using the IQ3_XXS quant which supposedly matches baseline, it feels significantly degraded." ([source](https://news.ycombinator.com/item?id=49957317))
- Only moderately intelligent (3 comments): "I've tried it but so far I have not seen anything that overly impressed me in terms of accuracy, though the speed is definitely there." ([source](https://news.ycombinator.com/item?id=49957786)) "Their "next" variants are usually undercooked, but useful for the community to verify support for inference stacks." ([source](https://news.ycombinator.com/item?id=49436116)) "it's just moderately intelligent - if you observe the CoT while reversing, you'll find that struggles" ([source](https://news.ycombinator.com/item?id=49910798))

## Alternatives hackers recommend

- [Qwen 3.8 27B](https://hackerapproved.com/p/qwen-3-8-27b.md): 1173 recommend
- [Z.ai GLM](https://hackerapproved.com/p/glm-5-2.md): 1043 recommend
- [Kimi](https://hackerapproved.com/p/kimi-k3.md): 931 recommend
- [Gemma 4](https://hackerapproved.com/p/gemma-4.md): 713 recommend
- [Qwen](https://hackerapproved.com/p/qwen.md): 474 recommend

## Cite as

Hacker Approved, "Qwen 3.8 Flash Next", https://hackerapproved.com/p/qwen-3-8-flash-next/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).