geekradar
Int
Explore GeekRadar

More to explore

Good gear All topics About us Say hello
GeekRadar / The comparison desk
COMPARE.
AI models
alibaba

Qwen 3.5 Flash (hosted 35B-A3B)

AI model Released

Capability & preference

General capability · ECI

A cross-benchmark capability estimate. Index points, not a percentage; intervals show uncertainty.

Qwen 3.5 Flash (hosted 35B-A3B) 143.8* Qwen 3.5 Flash (hosted 35B-A3B) · 95% interval 141.8–145.8

Model performance

Radar task score: four tests, equal weight. An experimental summary of these tasks, not a universal intelligence rating.

Qwen 3.5 Flash (hosted 35B-A3B)

Radar task score

33.0* / 100

4/4 core tests · experimental v1

5 benchmark results for this configuration

Scientific reasoning

GPQA Diamond · 25%

Expert-written science questions; strict multiple-choice scoring. Chance accuracy is 25%.

Qwen 3.5 Flash (hosted 35B-A3B) 82.3*% Approx. 95% interval: 77.0–87.7%

Novel problem solving

Mystery Game Puzzles · 25%

Best-next-move puzzles in an undisclosed game variant. A narrow reasoning test, not general intelligence.

Qwen 3.5 Flash (hosted 35B-A3B) 20.0*% Approx. 95% interval: 12.1–27.9%

Advanced mathematics

FrontierMath Tiers 1-3 · 25%

Private v2 mathematics problems with Python tools. Earlier versions and Tier 4 are excluded.

Qwen 3.5 Flash (hosted 35B-A3B) 9.5*% Approx. 95% interval: 6.1–12.9%

Factual answers

SimpleQA Verified · 25%

Correct factual answers under a guessing prompt, graded by a model. Not a hallucination or abstention score.

Qwen 3.5 Flash (hosted 35B-A3B) 20.3*% Approx. 95% interval: 17.8–22.8%

Competition mathematics

OTIS Mock AIME 2024-2025 · Separate result

Competition problems with model-extracted answers checked against an exact key. Earlier grading procedures are excluded.

Qwen 3.5 Flash (hosted 35B-A3B) 84.4*% Approx. 95% interval: 73.7–95.2%

Bars show reported task scores. Whiskers are approximate 95% intervals where standard errors are available. Different reasoning budgets are not equal-cost tests. Missing results are never zero.

Run it locally

Inference · memory estimate

Qwen 3.5 Flash (hosted 35B-A3B)

Not enough verified architecture information for a reliable VRAM estimate.

System configuration

Detailed specifications

Specifications · 2 fields
Organization
Alibaba*
Organization country
China*

* Provisional data