alibaba

Qwen 3.6 72B

Released 2026-04-05Context 256Kqwen3-6-72b

PipelineScore

80.1MAINLINE

Ranked #17 of 52 models · 67th percentileAn even spread: no standout, no liability (78 to 82 across all five categories). Best-fit profile: Local-first.

Category breakdown

Score per category, normalized 0–100 against the v1 anchor.

Code

79.4

Reason

81.0

Tool Use

78.3

RAG

78.9

Speed

82.1

Strengths

Speed82.1

Reason81.0

Code79.4

Same model, different rigs

Every submission of Qwen 3.6 72B on the 0–100 scale. The spread is the point: where it runs changes what you get.

0255075100

Best 81.6 on untagged · lowest 79.1 on cloud-api · spread 2.5 pts across 12 runs. Hover a dot for its rig.

Sample tasks

A taste of what the test pack measures. Full prompts are private and rotated daily.

CodeDifficulty 1code-fib-1

Fibonacci function

Write a Python `fib(n)` returning the nth Fibonacci number, O(n).

ReasonDifficulty 1reason-math-1

Train meeting time

Two trains, opposite directions, given speeds and start times — when do they meet?

RAGDifficulty 2rag-extract-1

Extract metrics to JSON

From the context, extract net sales, operating margin, and free cash flow as a JSON object. Numbers only.

Tool UseDifficulty 2tool-schema-1

OpenAPI param selection

Given an OpenAPI schema with limit/offset/sort, fill JSON for 'next 50, recent first.'

RAGDifficulty 2rag-grounding-1

Refuses to fabricate

Context lacks the answer — does the model fabricate or correctly say it can't?

Compare with

Qwen 3.6 72B vs openai/gpt-oss-20b Qwen 3.6 72B vs DeepSeek R1 671B-A37B Qwen 3.6 72B vs Qwen 3 235B-A22B MoE Qwen 3.6 72B vs DeepSeek V3 671B-A37B Qwen 3.6 72B vs DeepSeek Coder V2 236B