Crosshair
Leaderboard
Language ModelsOpen weights

Qwen3.6-27B

Alibaba's dense open-weights Qwen3.6 (Apache-2.0); multimodal.

textimagecodeOfficial site
Crosshair Index
60.1
#16 of 32 · Language Models
Provider
Alibaba Qwen
Released
2026-04-22
Parameters
27B
Context
262.144K tokens

Token pricing

Input
$0.25 /1M
Output
$1.49 /1M

USD per 1M tokens · cache read = cached input (hit), cache write = caching surcharge · official list pricing (June 2026).

Industry skill web

Professional-domain strengths, composed from the benchmarks relevant to each field. Highlight an axis to see the benchmarks behind it.

SoftwareIB / FinanceLawMedicineResearchConsultingAccounting

Investment Banking

83
skill

Financial analysis over filings and credit agreements — valuation math, document QA, and the quantitative reasoning behind deals.

Corporate Finance62.3%
MMLU-Pro86.2%
GPQA Diamond87.8%

Scorecard

BenchmarkScoreSourceStatus
GPQA Diamond
Reasoning
87.8%Alibaba — Qwen3.6-27B model cardvendorunverified
Humanity's Last Exam
Frontier
24%Alibaba — Qwen3.6-27B model cardvendorunverified
SWE-bench Verified
Agentic Coding
77.2%Alibaba — Qwen3.6-27B model cardvendorunverified
LiveCodeBench
Coding
83.9%Alibaba — Qwen3.6-27B model cardvendorunverified
MMLU-Pro
Knowledge
86.2%Alibaba — Qwen3.6-27B model cardvendorunverified
LMArena Elo
Human Preference
—not evaluated
AA Intelligence Index
Composite
46Artificial Analysis3rd-partyunverified
Corporate Finance
Finance
62.3%Vals AI — CorpFin v23rd-partyunverified
LegalBench
Law
—not evaluated
TaxEval
Tax & Accounting
71.3%Vals AI — TaxEval v23rd-partyunverified
Medical Coding
Medicine
—not evaluated