Noometry

Alibaba (Qwen), open weights

Qwen3.5 27B

Qwen3.5 27B by Alibaba (Qwen) ranks 127th of 354 ranked models on the Noometry Index as of October 2026, with a score of 41.9. Its strongest category is multimodal, where it ranks 59th. API pricing starts at $0.30 per million input tokens and $2.40 per million output tokens, with a 262K-token context window.

Last verified

Noometry rank
#127 of 354
Index score
41.9
Evidence
Confirmed 28 results
Provider
Alibaba (Qwen)
Released
February 23, 2026
Weights
Open weights
Reasoning
Yes
Context window
262K
Max output
66K
Input price
$0.30 / M
Output price
$2.40 / M
Blended price
$0.82 / M
Output speed
Not measured
Value
#94 of 219
Knowledge cutoff
Unknown
Input
text, image, video, audio
Hugging Face
Qwen/Qwen3.5-27B

Category scores

Each category score combines every public result we have in that category.

  1. Coding 38.9
  2. Reasoning 27.5
  3. Math 38.8
  4. Knowledge 38.0
  5. Multimodal 39.4
  6. Multilingual 50.8
  7. Instruction Following 73.5
  8. Long Context 43.1
  9. Writing & Preference 59.3
CategoryScoreRankResults
Coding38.9#1683
Reasoning27.5#1175
Math38.8#1272
Knowledge38.0#1502
Multimodal39.4#591
Multilingual50.8#1151
Instruction Following73.5#1191
Long Context43.1#1061
Writing & Preference59.3#1113

Strengths and weaknesses

Categories where Qwen3.5 27B places highest and lowest among the models ranked in each, with its score against that category's median.

Strongest categories

CategoryScorevs medianRank
Reasoning27.5+3.9#117 of 350, top 34%
Writing & Preference59.3+5.6#111 of 312, top 36%
Long Context43.1+2.2#106 of 296, top 36%

Weakest categories

CategoryScorevs medianRank
Coding38.9+0.1#168 of 340, top 50%
Knowledge38.0+0.7#150 of 314, top 48%
Multimodal39.4+0.9#59 of 128, top 47%

Closest competitors

The models ranked just above and below Qwen3.5 27B. When scores are this close, price and speed are often the better way to choose.

ModelRankScoreBlended $/MSpeed
Qwen3.5 35B-A3B#12342.0$0.69—Compare
GLM-4.7#12442.0$1—Compare
GPT-5.4 nano#12541.9$0.4619Compare
Amazon Nova Experimental Chat 10 09#12641.9——Compare
GPT-5 Mini#12841.8$0.693Compare
ERNIE 5.0 0110#12941.8——Compare
Granite 4.2 30b#13041.8——Compare
Muse Glimmer#13141.7——Compare

Sponsored placements are available on pages like this one. Advertise on Noometry

Benchmark results

Every published result we track, with its source. Bold rows are the ones used for ranking; where several exist we prefer independent runs over self-reported numbers.

Coding

BenchmarkScorePositionSettingSourceDate
LMArena WebDev1358#86 of 113, top 77%LMArena2026-10-08
WeirdML39.5%#78 of 119, top 66%Epoch AI
LMArena Coding1427#114 of 294, top 39%LMArena2026-10-08
ALE-Bench349.45#93 of 105, top 89%Epoch AI

Agentic & Tool Use

BenchmarkScorePositionSettingSourceDate
Vending-Bench 2201.98#52 of 60, top 87%Epoch AI

Reasoning

BenchmarkScorePositionSettingSourceDate
NYT Connections (extended)47.9%#63 of 91, top 70%Lech Mazur benchmarks
Thematic Generalization45.5%#18 of 23, top 79%Lech Mazur benchmarks
LMArena Hard Prompts1414#113 of 297, top 39%LMArena2026-10-08
DTBench82.4%#65 of 151, top 44%Epoch AI
LMCA34%#66 of 125, top 53%Epoch AI

Math

BenchmarkScorePositionSettingSourceDate
MathArena Final-Answer Competitions56.7%#25 of 29, top 87%MathArena
LMArena Math1429#88 of 285, top 31%LMArena2026-10-08

Knowledge

BenchmarkScorePositionSettingSourceDate
Vectara Hallucination Rate (lower is better)12.1%#75 of 96, top 79%Vectara Hallucination Leaderboard
LMArena Expert1428#98 of 273, top 36%LMArena2026-10-08

Multimodal

BenchmarkScorePositionSettingSourceDate
LMArena Vision1241#63 of 122, top 52%LMArena2026-10-09

Multilingual

BenchmarkScorePositionSettingSourceDate
LMArena Non-English1390#115 of 297, top 39%LMArena2026-10-08
LMArena Chinese1478#66 of 285, top 24%LMArena2026-10-08
LMArena French1410#113 of 223, top 51%LMArena2026-10-08
LMArena German1393#99 of 231, top 43%LMArena2026-10-08
LMArena Japanese1345#101 of 211, top 48%LMArena2026-10-08
LMArena Korean1358#98 of 213, top 47%LMArena2026-10-08
LMArena Russian1390#119 of 283, top 43%LMArena2026-10-08
LMArena Spanish1407#102 of 226, top 46%LMArena2026-10-08

Instruction Following

BenchmarkScorePositionSettingSourceDate
LMArena Instruction Following1393#111 of 298, top 38%LMArena2026-10-08

Long Context

BenchmarkScorePositionSettingSourceDate
LMArena Longer Query1413#102 of 291, top 36%LMArena2026-10-08

Writing & Preference

BenchmarkScorePositionSettingSourceDate
LMArena Text1409#111 of 297, top 38%LMArena2026-10-08
LMArena Creative Writing1362#119 of 295, top 41%LMArena2026-10-08
LMArena Multi-Turn1410#110 of 295, top 38%LMArena2026-10-08

API pricing by provider

RouteInput $/MOutput $/MCached input $/MChecked
alibaba$0.30$2.40—2026-10-10
deepinfra$0.26$2.60—2026-10-10
openrouter$0.26$2.60—2026-10-10

All Alibaba (Qwen) API prices → Estimate your cost →

Compare Qwen3.5 27B

Other Alibaba (Qwen) models

Frequently asked questions

How good is Qwen3.5 27B?

Qwen3.5 27B by Alibaba (Qwen) ranks 127th of 354 ranked models on the Noometry Index as of October 2026, with a score of 41.9. Its strongest category is multimodal, where it ranks 59th. API pricing starts at $0.30 per million input tokens and $2.40 per million output tokens, with a 262K-token context window.

How much does Qwen3.5 27B cost?

Qwen3.5 27B costs $0.30 per million input tokens and $2.40 per million output tokens on Alibaba (Qwen)'s own API.

What is Qwen3.5 27B's context window?

Qwen3.5 27B accepts up to 262K tokens of input and can write up to 66K tokens in one response.

Is Qwen3.5 27B open source?

Yes. Qwen3.5 27B's weights are downloadable from Hugging Face (Qwen/Qwen3.5-27B); check the license for commercial terms.

What are Qwen3.5 27B's strengths and weaknesses?

Relative to other ranked models, Qwen3.5 27B places best in reasoning, writing & preference, long context and lowest in coding, knowledge, multimodal.

What is Qwen3.5 27B best at?

Its best category is multimodal, where it ranks 59th on Noometry.