vibecodingnav
中/EN
Models/qwen3-8-max

Qwen3.8 Max

Two preliminary WebDev rows, 1674 and 1672. Sixth on Android Bench 1.0 at 87.0%.

Alibaba · Proprietary · Context 1M · $2 / $6 per million tokens · Fetched 2026-10-10

LMArena WebDev
1674
#10 / 142 · ±8
LMArena Image-to-WebDev
1638
#9 / 64 · ±18
Android Bench
14.0%
#5 / 10 · 4.7–24.0

Benchmark scores

Boards older than 90 days are not counted as recommendations
BenchmarkSettingScoreRankCostAs of
LMArena WebDev09021674±8#10—2026-10-08
LMArena Image-to-WebDev09021638±18#9—2026-10-09
Android Benchqwen-coder14.04.7–24.0#5$260.22026-10-06
Terminal-Bench 4.0max · Claude Code27.0±3.8#25$4.6k2026-10-06
The Slop Index—64#11—2026-10-11
sudo L7xHigh reasoning15.6#6—2026-10-09
ReactBenchxhigh36.431.5–41.3#15$1.622026-10-10
DeepSWE v1.1xhigh · mini-swe-agent57.554.8–60.1#16$3.732026-09-22
SWE-bench Verified—This board has stopped updating——2026-02-26
Aider Polyglot—This board has stopped updating——2025-11-20

Use it or not

Why
  • Still in the Android top ten, cheaper than the leader.
When not
  • Both WebDev rows are marked Proprietary, and both are still preliminary.

Appears in these scenarios

Collected from scenario boards
Build a Web frontend#6sudo L7 15.6%Build an Android app#5Android Bench 14.0%Write CLI, scripts, and DevOps#16DeepSWE 57.5%Write docs & copy#11Slop Index 64

Peers