Advertise here
#102

Qwen3.5-27B

Qwen Release: 2026-02-24 Tested on: 2026-10-01 01:06 qwen/qwen3.5-27b::medium
27BDenseOpen source
(medium) (none)

Summary

Qwen3.5-27B scores 7.9 on AI BENCHY and ranks #102. It has 10.0 reliability, a 76.8% pass rate, $1.135 total cost, and 114.73s average response time.

What makes Qwen3.5-27B unique: It stands out most in Domain specific, where it ranks #4, while Coding is its weakest area at #13. It uses unusually many reasoning tokens, which can help explain its slower or more expensive runs.

Model facts

Researched on 2026-08-12

Reported
Parameters
27B
Architecture
Dense
Availability
Open source
License
Apache-2.0
Parameter source Availability source

Score

7.9

Consistency

7.9

Reliability

10.0

Total Cost (Current Price)

$1.135

Total Output Tokens

635,288

Total Input Tokens

288,919

Input Price

$0.195 / 1M

Output Price

$1.560 / 1M

Cache Read Price

N/A

Cache Write Price

N/A

Cache prices apply to input tokens. Reads reuse cached prompts; writes store them and can cost extra. Output tokens use the output price.

Tests Correct

Wrong Tests: 9

Attempt pass rate: 76.8%

Flaky tests

6

Flaky tests had mixed outcomes across runs (at least one pass and one fail).

Response Time (avg)

114.73s

Response Time (max): 1026.43s

Response Time (total): 2638.74s

Wrong answer: 4 Did not follow instructions: 2 Extra formatting: 1 Invalid tool call: 1 Timed out: 1

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#102 Qwen3.5-27B

medium
Cost
$0.008
Time
62.0s
Tokens
3,099 tok

Run history

Tested on Score Reliability Tests Correct Total Cost Compare
2026-10-01 01:06 Re-test 7.9 10.0 $1.135 Current run
2026-07-16 22:51 New test added 7.4 10.0 $0.981 ↓ Compare
2026-06-04 13:40 New test added 7.8 10.0 $0.536 ↓ Compare
2026-05-22 00:04 Suite changed 7.9 10.0 $0.607 Compare
2026-04-11 01:44 First recorded run 8.4 N/A $0.497 Compare

This run used a different benchmark suite. Keep suite changes in mind when reading historical movement.

Run comparison

RunBenchmark coverageScoreConsistencyReliabilityTests CorrectFlaky testsTotal Output TokensTotal Input TokensTotal CostResponse Time (avg)
2026-10-01 01:06 · Current run69/69 attempts7.97.910.014/236635,288288,919$1.135114.73s
2026-05-22 00:04 · Suite changed60/60 attempts7.98.910.013/203314,5830$0.60760.85s
Difference—0.0-1.00.0+1+3+320705+288919+$0.529+53882ms

Benchmark coverage differs: 69/69 attempts (Target: 3 repeats per test) versus 60/60 attempts (Target: 3 repeats per test). Totals and repeat-sensitive metrics are not directly comparable.

These two runs used different benchmark suites, so the deltas reflect both model changes and suite changes.

Charts

Choose the first model, then click a second model to open a side-by-side page.

Total Output Tokens

Score vs Total Output Tokens

Quick Compare

Qwen3.5-27BmediumvsGPT-5.6 TerralowQwen3.5-27BmediumvsGLM 5.3 FlashXmaxQwen3.5-27BmediumvsQwen3.8 Flash NextlowQwen3.5-27BmediumvsClaude Fable 5.1highQwen3.5-27BmediumvsGemini 3.8 FlashmediumQwen3.5-27BmediumvsGPT-5.4 MinimediumQwen3.5-27BmediumvsGPT-5.4 NanomediumQwen3.5-27BmediumvsQwen3.7 MaxnoneQwen3.5-27BmediumvsNemotron 3 UltramediumFree AvailableQwen3.5-27BmediumvsGemini 3.5 Flashmedium

Category Breakdown

Category Score Consistency Tests Correct
Agentic 10.0 10.0
Anti-AI Tricks 8.7 7.9
Coding 6.2 7.1
Combined 7.3 5.8
Data parsing and extraction 10.0 10.0
Domain specific 5.9 7.2
General Intelligence 6.1 3.1
Instructions following 10.0 10.0
Puzzle Solving 8.2 7.7
Tool Calling 10.0 10.0
Trivia 3.0 10.0

Compared models