All tasks / Models

Qwen3.8-Flash-Next

Qwen's open-weights fast MoE (125B total, 6B active, plus 51B n-gram embedding) under the Qwen community licence.

Alibaba (Qwen)
Open weights
Released 24 Aug 2026

Where it ranks

B
Chat & assistants
Models · Local · 40 on Artificial Analysis Intelligence Index (secondary benchmark)
B
Coding
Models · Local · 1,629 on LMArena WebDev / Code Arena (overall)
–
Image understanding
Models · Local · no result on this task’s benchmarks yet

Facts

Vendor: Alibaba (Qwen)
Family: Qwen Flash
Released: 24 Aug 2026
Weights: Open: download and run it yourself
License: qwen-community-1.0
Parameters: 180B (6B active)
GPU memory to run locally: ~104 GB (4-bit ≈ 180 B stored params (125 B MoE + 51 B n-gram embedding + 4 B MTP) × 0.57 GB + 1.5 GB ≈ 104 GB, estimate.)
Access: Free (open weights)
Last verified: 19 Oct 2026

Price from huggingface.co, 19 Oct 2026.

Benchmark results

Benchmark	Result	Entry on the leaderboard	As of	Source
Artificial Analysis Intelligence Index	40	`Qwen3.8-Flash-Next` Raw value from the page's embedded model data; the site displays it rounded to an integer.	19 Oct 2026	artificialanalysis.ai
LMArena WebDev / Code Arena (overall)	1,629	`Qwen3.8-Flash-Next` From LMArena's official leaderboard-dataset (webdev = Code Arena, split latest, category overall, published 2026-10-18), which mirrors https://arena.ai/leaderboard/webdev. Rank 20; 8278 votes; 95% CI 1621.0-1637.0.	18 Oct 2026	huggingface.co

Where you can use it

No tracked tool lists this model yet; as open weights it runs in local runtimes.

Sources