Which is better, Claude Opus 4.6 or Qwen3.6 27B?

Across 46 shared benchmarks tracked on Vector Wire, Claude Opus 4.6 leads on 34 and Qwen3.6 27B leads on 12 — Claude Opus 4.6 leads on more. Strengths differ by axis; the table below shows every head-to-head number with provenance.

SEEDRefresh Models —Deals —Regulatory —Sources —

Claude Opus 4.6 vs Qwen3.6 27B

AnthropicvsAlibaba46 shared benchmarks34–12 head-to-head

Benchmark	Claude Opus 4.6	Qwen3.6 27B
AA Agentic Index	67.6	27.5
AA Intelligence	44.9	37.7
AA-LCR	74.3	73.3
AA-Omniscience	13.7	-20
aime_2026	96.7	94.1
AndroidWorld	62	70.3
Artificial Analysis Coding Index	48.1	53.7
baby_vision	14.8	28.9
charxiv_rq	69.1	78.4
ClawEval-MM	54.7	50.4
critpt	12.6	1.1
ERQA	40.8	62.5
gdpval	55.9	32
GPQA Diamond	91.3	87.8
HLE	53.1	24
hmmt_feb_2026	96.2	84.3
hmmt_nov_2025	96.3	90.7
IFBench	62.5	69.1
imo_answer_bench	75.3	80.8
livebench_agentic_coding	49	39.3
livebench_coding	78.2	71.8
livebench_data_analysis	69.9	70.4
livebench_instruction_following	63.3	53.2
livebench_language	83.3	63.3
livebench_math	89.3	79.9
livebench_reasoning	88.7	70.3
LiveCodeBench v6	88.8	83.9
mathvision	71.2	85.1
MMLU-Pro	89.1	86.2
MMMU-Pro	77.3	75.8
nl2repo	49.8	36.2
OmniDocBench 1.5	86.6	89.4
OmniScience Accuracy	47	19.6
OmniScience Non-Hallucination	37.2	50.7
OSWorld-Verified	72.7	63.9
RealWorldQA	73.9	84.1
scicode	52	39.8
SWE-bench Multilingual	77.8	71.3
SWE-bench Pro	57.3	53.5
SWE-bench Verified	80.8	77.2
SWE-MM	27.1	25.7
Terminal Bench 2.1 (Terminus)	78.2	63.4
Terminal-Bench 2.0	65.4	59.3
Terminal-Bench Hard	65.4	34.8
τ²-Bench Telecom (AA run)	99.3	94.2
τ³-Bench	72.4	15.3

Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing.