Qwen3.5 consumes 14x more tokens than Gemma 4 for similar mobile scores
ArtificialAnlys · x · 2026-08-25
Achieving a score of 60 can cost 5M or 75M output tokens depending on the model. Qwen3.5 9B (Reasoning) spends 74.5M tokens and exceeds the 16K window on 29% of generations, while Gemma 4 E4B (Reasoning) reaches a similar score on just 5.2M tokens without hitting the limit. On mobile phones, token usage translates directly into time, energy, and heat, which are less tolerable than on servers.
Related event: Artificial Analysis and Liquid AI Launch On-Device Small Model Benchmark(9 posts)→
More from Models
- Unverified: internal snapshot labeled gpt-6-astra-api-priv-ev3 spotted on model tracker — lyraxana · 2026-10-16
- Muse Code 1.4.8 cuts token usage; all user quotas reset immediately — alexandr_wang · 2026-10-16
- Alibaba and Baidu to power key parts of Apple's AI services in China — nordicinst · 2026-10-16
- Cornell and Amazon revisit the 16x pretraining compute gap: diffusion LMs need only 3-8x — LucaAmb · 2026-10-16
- Image Decision Benchmark JevImageBench hits v0.3.0 with reshuffled Top 5 — airesearch12 · 2026-10-16
- Forecast: Meta's Muse Spark Open May Become the Best Non-Chinese Open-Weight LLM on Agentic Benchmarks — zsakib_ · 2026-10-16