Alibaba Cloud / Qwen Team · apache_2_0

Qwen3.5-27B

Qwen3.5-27B by Alibaba Cloud / Qwen Team appears in 1 source with Reasoning at 41.95. Best read for Code quality, Low cost, Multimodal.

CreatorAlibaba Cloud / Qwen Team

Release date2026-02-24

Knowledge cutoffNot published

Context262K tokens

Input price$0.3/M tokens

Output price$2.4/M tokens

Modalitytext + vision

CountryCN

Metrics

All source-backed metrics

Open in leaderboard

LLM Stats Rank68ranking · source_rank

Reasoning41.95reasoning · index_reasoning

Math42.35math · index_math

Coding21.35coding · index_code

Research20.44research · index_search

Writing22.23writing · index_communication

Vision29.36multimodal · index_vision

Tool calling12.68tool_calling · index_tool_calling

Long context30.92long_context · index_long_context

Finance45.62domain · index_finance

Legal45.73domain · index_legal

Healthcare43.86domain · index_healthcare

GPQA85.5 %reasoning · gpqa_score

SWE-bench Verified72.4 %coding · swe_bench_verified_score

Code Arena784.65 %coding · coding_arena_score

Humanity Last Exam48.5 %reasoning · hle_score

MMMLU85.9 %reasoning · mmmlu_score

MMMU82.3 %multimodal · mmmu_score

MMMU-Pro75 %multimodal · mmmu_pro_score

BrowseComp61 %research · browsecomp_score

CharXiv-R79.5 %multimodal · charxiv_r_score

ScreenSpot Pro70.3 %multimodal · screenspot_pro_score

Context262,144 tokenscontext · context

Speed100.96 c/sperformance · throughput

Latency39,283.26 msperformance · latency

Input price0.3 $/Mpricing · input_price

Output price2.4 $/Mpricing · output_price

Parameters27,000,000,000 paramsmodel · params

Evidence

Citations and source overlap

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 Qwen3.5-27B 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.