Qwen

90 models tracked · known for Qwen3-0.6B, Qwen3-4B, Qwen3-8B

Models tracked

312

Benchmark scores

100

Best score

18%

Reproduced

57 of 312

What the record says

Claims drawn from cited facts, not live model generation.

4 claims

Lab Summary

Qwen is the lab summarized here. It keeps a tracked blog that publishes recent posts, without the provided description indicating a narrower research, product, safety, or policy focus.

3 cited facts

Portfolio

Qwen has shipped 90 models and has 44 recent blog announcements. That mix signals an actively communicating lab with broad model output, rather than a quieter public presence.

4 cited facts

Evidence mix

312 benchmark scores across this lab’s models, tagged by evidence status.

reproduced57self-reported57unverified198

Models

Status

Sort

Model	Best score	Claim status
Qwen3.8 Max (0902)	100	unverified
Qwen 3.8 Max	99.44	independently reproduced
Qwen3-Max	97.13	independently reproduced
Qwen3.7-Max	95.55	independently reproduced
Qwen 3.6 Plus	93.33	independently reproduced
Qwen3.7-Plus	93.33	independently reproduced
Qwen2.5-72B	92.67	independently reproduced
Qwen 3.6 Max (Preview)	91.1	independently reproduced
Qwen2.5-Coder-32B	91.1	self-reported
Qwen3.6 27B	91.1	unverified
Qwen3.5 397B-A17B	88.88	unverified
Qwen2.5-Coder-14B	88.7	self-reported
Qwen 3.8 27B	87.5	unverified
Qwen2.5-Coder (7B)	86.7	self-reported
Qwen 3.5 Plus (hosted 397B-A17B)	86.65	independently reproduced
Qwen 3.6 35B-A3B	86.65	independently reproduced
Qwen3-235B-A22B-Thinking (Jul 2025)	86.65	independently reproduced
Qwen3.7 Flash	86.65	unverified
Qwen 3.5 Flash (hosted 35B-A3B)	84.43	independently reproduced
Qwen 3.6 Flash	84.43	independently reproduced
Qwen3-235B-A22B	83	independently reproduced
Qwen-14B	79.2	self-reported
Qwen3.5-35B-A3B	77.95	unverified
Qwen2-72B	76.53	independently reproduced
Qwen2.5-Coder-3B	75.7	self-reported
Qwen3-30B-A3B	75.3	unverified
Qwen3-32B	74.2	unverified
Qwen2.5-Max	72.9	independently reproduced
Qwen3.5-9B	71.97	unverified
Qwen3-30B-A3B-Thinking (Jul 2025)	70.25	unverified
Qwen-7B	67.9	self-reported
Qwen3-14B	66.36	unverified
Qwen2.5-Coder (1.5B)	65.8	self-reported
Qwen3-235B-A22B-Instruct (Jul 2025)	64	unverified
Qwen2.5-7B	63.87	unverified
Qwen3-30B-A3B-Instruct (Jul 2025)	62.18	unverified
Qwen3-8B	62.1	unverified
Qwen-1_8B	58.4	self-reported
Qwen2.5-32B	56.07	unverified
CodeQwen1.5-7B	37.7	self-reported
Qwen2.5-Coder-0.5B	34.5	self-reported
Qwen3-0.6B	—	unverified
Qwen3-4B	—	unverified
Qwen3-8B	—	unverified
Qwen3-14B	—	unverified
Qwen2.5-7B-Instruct	—	unverified
Qwen2.5-0.5B-Instruct	—	unverified
Qwen2.5-1.5B-Instruct	—	unverified
Qwen2.5-3B-Instruct	—	unverified
Qwen3-1.7B	—	unverified
Qwen3-32B	—	unverified
Qwen3-4B-Instruct-2507	—	unverified
Qwen3-8B-AWQ	—	unverified
Qwen2.5-Coder-14B-Instruct	—	unverified
Qwen2.5-Coder-7B-Instruct	—	unverified
Qwen2.5-14B-Instruct	—	unverified
Qwen2.5-32B-Instruct	—	unverified
Qwen2.5-0.5B	—	unverified
Qwen3-30B-A3B	—	unverified
Qwen3-Coder-30B-A3B-Instruct	—	unverified
Qwen3-30B-A3B-Instruct-2507	—	unverified
Qwen3-Coder-Next	—	unverified
Qwen2.5-Coder-32B-Instruct	—	unverified
Qwen2.5-7B	—	unverified
Qwen3-8B-Base	—	unverified
Qwen2.5-Coder-7B-Instruct-AWQ	—	unverified
Qwen3-0.6B-Base	—	unverified
Qwen2.5-1.5B	—	unverified
Qwen2.5-3B-Instruct-AWQ	—	unverified
Qwen3-4B-Thinking-2507	—	unverified
Qwen3-8B-GGUF	—	unverified
Qwen3-1.7B-Base	—	unverified
Qwen3-4B-Base	—	unverified
Qwen3-4B-GGUF	—	unverified
Qwen-72B	—	unverified
Qwen3-Coder-480B-A35B-Instruct-FP8	—	unverified
Qwen2-1.5B-Instruct	—	unverified
Qwen2.5-72B-Instruct	—	unverified
Qwen2.5-Coder-7B	—	unverified
Qwen2-0.5B	—	unverified
Qwen2.5-Coder-3B-Instruct	—	unverified
Qwen1.5-MoE-A2.7B	—	unverified
Qwen3-14B-Base	—	unverified
Qwen3-30B-A3B-Instruct-2507-FP8	—	unverified
Qwen2.5-Coder-1.5B-Instruct	—	unverified
Qwen2.5-Math-1.5B	—	unverified
Qwen2.5-32B-Instruct-GPTQ-Int4	—	unverified
Qwen2-7B-Instruct	—	unverified
Qwen3.8-27B	—	unverified

Benchmarks led

Where a model from Qwen holds the top score, with the benchmark’s integrity grade.

OTIS Mock AIME 2024-2025
B

Source receipt

Source facts, citations, and refresh stamp for this record.

updated 2026-10-19

Source records: 375
Fact rows: 796
Fact types: 7
Sources: 4

Sources: curated_models, epoch_benchmarks, huggingface_models, lab_blog_rss

Qwen

What the record says

Lab Summary

Portfolio

Evidence mix

Models

Benchmarks led

Source receipt

Follow the record

Other labs

Explore the record

Qwen

What the record says

Lab Summary

Portfolio

Evidence mix

Models

Benchmarks led

Source receipt

Follow the record

Other labs

Explore the record