Which is better, Muse Spark 1.1 or Qwen3.5 4B?

Muse Spark 1.1 vs Qwen3.5 4B

15 SHARED BENCHMARKS

Across 15 shared benchmarks, Muse Spark 1.1 scores higher on 14 and Qwen3.5 4B on 1. The widest gap is HLE, where Muse Spark 1.1 scores 62.1 against 9.9. Qwen3.5 4B is the cheaper of the two on tracked API pricing ($0.03 against $1.25 per million input tokens).

METAVSALIBABA15 SHARED14–1 HEAD-TO-HEAD

BenchmarkMuse Spark 1.1MARGINQwen3.5 4B

AA Agentic Index27.536.3

AA Intelligence34.320

AA-LCR77.763

AA-Omniscience28.1-58.4

Artificial Analysis Coding Index71.322.6

critpt15.10

gdpval39.78.3

GPQA Diamond89.877.1

HLE62.19.9

multichallenge75.349

OmniScience Accuracy5215.1

OmniScience Non-Hallucination5013.4

scicode58.818.3

TauBench V3 - Banking31.86.8

Terminal-Bench 2.18025.8

Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing. Quoted rates are the price-setter row we currently track for each model — its direct or vendor-official listing where one exists (direct, direct), otherwise the lowest tracked offer.

Compare Muse Spark 1.1 withALL PAIRINGS →

MetaLlama 3.1 Instruct 405B

Across the fieldGemini 3.6 Flash Nemotron 3 Super 120B A12B GPT-5.5 Claude Opus 4.8 GPT-4o GPT-5.6 Sol Gemini 3.1 Pro GPT-5.4

Compare Qwen3.5 4B withALL PAIRINGS →

AlibabaQwen3.5 35B A3B Qwen3 VL 4B (Reasoning)Qwen3.5 397B A17B Qwen2.5 Instruct 72B Qwen3.5 27B Qwen3.6 Plus

Across the fieldGLM-5.1 GLM-5.2 o4-mini Claude Sonnet 4.6 o1 Gemma 4 31B Kimi K2 Instruct Gemma 4 E2B