modelgrep

Google: Gemini 3.5 Flash-Lite vs Qwen: Qwen3.6 27B

Google: Gemini 3.5 Flash-Lite wins on more metrics (5 of 8), but the right pick depends on what you optimize for — see the breakdown below.

MetricGoogle: Gemini 3.5 Flash-LiteQwen: Qwen3.6 27B
Intelligence Index36.537.1
Coding Index49.353.7
GPQA Diamond84%84%
Design Arena Elo
Speed (tokens/sec)341
Latency880ms
Input price /M$0.300$0.450
Output price /M$2.50$2.70
Context window1.0M262K
CapabilitiesReasoningToolsJSONVisionAudioReasoningToolsJSONVision