Noometry

Model comparison

Granite 4.0 H Small vs Qwen3.5-Flash

Qwen3.5-Flash is the stronger model overall, scoring 42.3 to 35.9 on the Noometry Index.

Last verified . 14 shared benchmarks.

Granite 4.0 H Small IBM

35.9

Rank #213 Confirmed

Qwen3.5-Flash Alibaba (Qwen)

42.3

Rank #105 Confirmed

Summary

  • They share 14 benchmarks with published results for both. Granite 4.0 H Small scores higher in 1 category and Qwen3.5-Flash in 7 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in writing & preference, where Qwen3.5-Flash leads 58.2 to 44.0.
  • The biggest single-benchmark swing is Vectara Hallucination Rate: 5.2% for Granite 4.0 H Small and 10.5% for Qwen3.5-Flash.
  • Granite 4.0 H Small has downloadable open weights; the other is API-only.

Side by side

Granite 4.0 H SmallQwen3.5-Flash
ProviderIBMAlibaba (Qwen)
Noometry Index35.942.3
Released—2026-02-23
WeightsOpenProprietary
Context window—1M
Max output—66K
Input $ / M tokens—$0.10
Output $ / M tokens—$0.40
Results tracked1932

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

Category by category

Coding Granite 4.0 H Small leads

Granite 4.0 H Small: 35.8 (#207), Qwen3.5-Flash: 33.9 (#234)

BenchmarkGranite 4.0 H SmallQwen3.5-Flash
LMArena Coding12471412
LMArena WebDev—1243
ALE-Bench—221.8

Agentic & Tool Use Not comparable

Granite 4.0 H Small: —, Qwen3.5-Flash: —

BenchmarkGranite 4.0 H SmallQwen3.5-Flash
Vending-Bench 2—462.69

Reasoning Qwen3.5-Flash leads

Granite 4.0 H Small: 23.4 (#164), Qwen3.5-Flash: 32.9 (#63)

BenchmarkGranite 4.0 H SmallQwen3.5-Flash
LMArena Hard Prompts12401404
Chess Puzzles—21%
Mystery Game Puzzles—20%
DTBench—82.9%
LMCA—29.1%
Epoch Capabilities Index—143.98

Math Qwen3.5-Flash leads

Granite 4.0 H Small: 29.3 (#219), Qwen3.5-Flash: 36.3 (#117)

BenchmarkGranite 4.0 H SmallQwen3.5-Flash
LMArena Math12511407
FrontierMath (Tiers 1-3)—18.2%
OTIS Mock AIME 2024-2025—84.4%
Omni-MATH29.6%—
FrontierMath (Feb 2025 set)—6.2%
FrontierMath Tier 4 (v1)—0%

Knowledge Qwen3.5-Flash leads

Granite 4.0 H Small: 31.5 (#210), Qwen3.5-Flash: 43.2 (#84)

BenchmarkGranite 4.0 H SmallQwen3.5-Flash
Vectara Hallucination Rate5.2%10.5%
LMArena Expert12441408
GPQA Diamond—82.3%
SimpleQA Verified—20.3%
MMLU-Pro56.9%—
GPQA (HELM)38.3%—

Multilingual Qwen3.5-Flash leads

Granite 4.0 H Small: 38.7 (#223), Qwen3.5-Flash: 50.7 (#116)

BenchmarkGranite 4.0 H SmallQwen3.5-Flash
LMArena Non-English12161385
LMArena Chinese12501446
LMArena Russian11981379
LMArena Spanish12581399
LMArena French—1412
LMArena German—1389
LMArena Japanese—1369
LMArena Korean—1344

Instruction Following Qwen3.5-Flash leads

Granite 4.0 H Small: 68.9 (#178), Qwen3.5-Flash: 72.8 (#133)

BenchmarkGranite 4.0 H SmallQwen3.5-Flash
LMArena Instruction Following12211374
IFEval89%—

Long Context Qwen3.5-Flash leads

Granite 4.0 H Small: 37.6 (#209), Qwen3.5-Flash: 42.4 (#119)

BenchmarkGranite 4.0 H SmallQwen3.5-Flash
LMArena Longer Query12401392

Writing & Preference Qwen3.5-Flash leads

Granite 4.0 H Small: 44.0 (#222), Qwen3.5-Flash: 58.2 (#117)

BenchmarkGranite 4.0 H SmallQwen3.5-Flash
LMArena Text12411397
LMArena Creative Writing12101343
LMArena Multi-Turn12431393
WildBench73.9%—

Frequently asked questions

Is Granite 4.0 H Small better than Qwen3.5-Flash?

Qwen3.5-Flash is the stronger model overall, scoring 42.3 to 35.9 on the Noometry Index.

Is Granite 4.0 H Small or Qwen3.5-Flash better for coding?

Granite 4.0 H Small scores higher on coding benchmarks: 35.8 versus 33.9 in the Noometry coding category.

How many benchmarks do Granite 4.0 H Small and Qwen3.5-Flash share?

14 benchmarks have published results for both models. Granite 4.0 H Small has 19 scored results on Noometry and Qwen3.5-Flash has 32.

Related comparisons

Go deeper