Alibaba Cloud / Qwen Team · apache_2_0

Qwen3.5-397B-A17B

Qwen3.5-397B-A17B by Alibaba Cloud / Qwen Team appears in 2 sources with Reasoning at 48.93. Best read for Code quality, Coding, High intelligence.

CreatorAlibaba Cloud / Qwen Team

Release date2026-02-16

Knowledge cutoffNot published

Context262K tokens

Input price$0.6/M tokens

Output price$3.6/M tokens

Modalitytext + vision

CountryCN

Metrics

All source-backed metrics

Open in leaderboard

LLM Stats Rank22ranking · source_rank

Reasoning48.93reasoning · index_reasoning

Math45.99math · index_math

Coding29.98coding · index_code

Research25.71research · index_search

Writing28.7writing · index_communication

Vision27.84multimodal · index_vision

Tool calling21.87tool_calling · index_tool_calling

Long context38.68long_context · index_long_context

Finance53.15domain · index_finance

Legal53.18domain · index_legal

Healthcare52.57domain · index_healthcare

GPQA88.4 %reasoning · gpqa_score

SWE-bench Verified76.4 %coding · swe_bench_verified_score

Code Arena1,294.63 %coding · coding_arena_score

Humanity Last Exam28.7 %reasoning · hle_score

MMMLU88.5 %reasoning · mmmlu_score

BrowseComp69 %research · browsecomp_score

Toolathlon38.3 %tool_calling · toolathlon_score

Context262,144 tokenscontext · context

Speed147.39 c/sperformance · throughput

Latency20,956.19 msperformance · latency

Input price0.6 $/Mpricing · input_price

Output price3.6 $/Mpricing · output_price

Parameters397,000,000,000 paramsmodel · params

Arena Rating1,441.64metric

Arena Rank36metric

Vote Count24,385metric

Evidence

Citations and source overlap

Arena AILM Arena text leaderboard (overall) mirrored in the public leaderboard dataset.Retrieved 2026-05-20 LLM StatsPublic full LLM leaderboard advanced explorer snapshot, including ranking, benchmark indexes, pricing, speed, context, license, and model metadata.Retrieved 2026-06-01

FAQ

How should I read this profile?

Treat this as a source-backed model dossier, not an EvalKit-run verification. The public values are replicated from linked sources and kept source-scoped.

Is Qwen3.5-397B-A17B verified by EvalKit?

No. EvalKit currently shows 0 verified rows until real run evidence exists.

Why can metrics disagree?

Different sources test different tasks, dates, prompts, and aggregation methods. EvalKit keeps those differences visible instead of merging them into a fake universal score.