priorsBuilder

Alibaba · released 2026-02-25

Qwen 3.5 Flash

Measured on 23 benchmarks.

General frontier

Not ranked in this mix — too little measured data for its benchmarks.

Your mix’s benchmarks

0 of 8 measured

This model has no results on your mix’s benchmarks.

All 23 benchmarks

show 23 more

coding

Arena WebDev

122 results
  1. #114max of 1221244

Leader: Opus 5.5 · 1813

coding

LiveCodeBench (Vals)

127 results
  1. #49max of 12783.3%

Leader: Fable 5.1 · 90.5%

coding

SWE-bench Verified (Vals)

85 results
  1. #71max of 8564.4%

Leader: Opus 5 · 97.0%

coding

Terminal-Bench 2.0 (Vals)

63 results
  1. #52max of 6324.7%

Leader: GPT-5.5 · 73.2%

composite

Arena Text

220 results
  1. #124max of 2201396

Leader: Gemini 4 · 1525

composite

Arena Text — Multi-Turn

220 results
  1. #128max of 2201396

Leader: Gemini 4 · 1553

economics

CorpFin (Vals)

117 results
  1. #45max of 11763.6%

Leader: Opus 5 · 73.2%

economics

MortgageTax (Vals)

84 results
  1. #24max of 8467.4%

Leader: Opus 5 · 72.1%

economics

TaxEval (Vals)

125 results
  1. #64max of 12572.2%

Leader: Muse Spark 1.2 · 80.4%

knowledge

CaseLaw (Vals)

53 results
  1. #31max of 5355.9%

Leader: Grok 4.3 · 79.3%

knowledge

GPQA Diamond (Vals)

122 results
  1. #60max of 12282.8%

Leader: Gemini 3.1 Pro · 95.5%

knowledge

MedCode

94 results
  1. #77max of 9433.0%

Leader: Opus 5 · 63.6%

knowledge

MedScribe

96 results
  1. #83max of 9670.6%

Leader: Opus 5.5 · 91.4%

knowledge

MMLU Pro (Vals)

122 results
  1. #65max of 12284.1%

Leader: Fable 5.1 · 92.4%

knowledge

MMMU Pro (Vals)

81 results
  1. #38max of 8181.9%

Leader: Fable 5.1 · 90.6%

math

AIME (Vals)

82 results
  1. #19max of 8292.5%

Leader: Gemini 3.1 Pro · 98.1%

math

FrontierMath Tiers 1–3 (v2)

115 results
  1. #92none of 11518.2%
  2. #103max of 1159.5%

Leader: GPT-6.1 Sol · 93.7%

reasoning

LegalBench

135 results
  1. #35max of 13584.3%

Leader: Fable 5 · 88.6%

reasoning

SAGE

80 results
  1. #52max of 8042.5%

Leader: Opus 4.7 · 56.1%

reliability

Tool-call error rate

171 results
  1. #159max of 1718.0%

Leader: GPT-5.4 Pro · 0.0%

reliability

Uptime (7d)

181 results
  1. #62max of 18199.9%

Leader: GPT-5 Pro · 100.0%

speed

Provider latency

177 results
  1. #26max of 1770.6s

Leader: Qwen 2.5 7B · 0.2s

speed

Provider throughput

176 results
  1. #49max of 17683.0

Leader: Muse Spark 1.1 · 204.3