GPT-5.4 nano (xhigh) vs Qwen3.5 397B A17B (Non-reasoning)

OpenAI · Alibaba · catalog 2026-09-14

Trade-off: GPT-5.4 nano (xhigh) is cheaper ($0.179 vs $0.9/M); Qwen3.5 397B A17B (Non-reasoning) scores higher on intelligence (21.4 vs 21.2).
GPT-5.4 nano (xhigh)Qwen3.5 397B A17B (Non-reasoning)
Speed (tok/s)154.2 measured81.7 measured
Blended $/M$0.179 derived$0.9 derived
Intelligence index21.2 measured21.4 measured
TTFT (ms)6427 measured224 measured
Context400k list—
Coding indexnot measured for both
Agentic indexnot measured for both

GPT-5.4 nano (xhigh) card · Qwen3.5 397B A17B (Non-reasoning) card · 3D observatory

Verdicts use price×intelligence dominance only, and only where both sides are measured; null cells stay null with their provenance (em-dash = not in catalog). Percentile-free by design — this page compares two models, not the field. Generated page — do not hand-edit.