Gemini 3.1 Flash-LitevsQwen3.6-35B-A3B
Across 3 shared benchmarks, Gemini 3.1 Flash-Lite leads overall: Gemini 3.1 Flash-Lite wins 3, Qwen3.6-35B-A3B wins 0, with 0 ties and an average score difference of +2.93.
Gemini 3.1 Flash-Lite
Google Deep Mind · 2026-05-07 · Multimodal model
Qwen3.6-35B-A3B
阿里巴巴 · 2026-04-16 · Reasoning model
Gemini 3.1 Flash-Lite3 wins(100%)(0%)0 winsQwen3.6-35B-A3B
Benchmark scores
Grouped by capability, sorted by largest gap within each. 3 shared benchmarks.
Coding and Software Engineer
Gemini 3.1 Flash-Lite 1/1| Benchmark | Gemini 3.1 Flash-Lite | Qwen3.6-35B-A3B | Diff |
|---|---|---|---|
| SciCode | 43.4095 / 130Thinking (No Tools) | 36.60112 / 130Thinking (No Tools) | +6.80 |
General Knowledge
Gemini 3.1 Flash-Lite 1/1| Benchmark | Gemini 3.1 Flash-Lite | Qwen3.6-35B-A3B | Diff |
|---|---|---|---|
| CritPt | 1.10147 / 200Thinking (No Tools) | 0.30182 / 200Thinking (No Tools) | +0.80 |
Multimodal Understanding
Gemini 3.1 Flash-Lite 1/1| Benchmark | Gemini 3.1 Flash-Lite | Qwen3.6-35B-A3B | Diff |
|---|---|---|---|
| GDP.pdf | 7.8088 / 118Thinking (No Tools) | 6.6090 / 118Thinking (No Tools) | +1.20 |
Specs
| Field | Gemini 3.1 Flash-Lite | Qwen3.6-35B-A3B |
|---|---|---|
| Publisher | Google Deep Mind | 阿里巴巴 |
| Release date | 2026-05-07 | 2026-04-16 |
| Model type | Multimodal model | Reasoning model |
| Architecture | Dense | MoE |
| Parameters | Not available | 35B |
| Context length | 1M | 200K |
| Max output | 64K | 80K |
Summary
- Gemini 3.1 Flash-Liteleads in:Coding and Software Engineer (1/1), General Knowledge (1/1), Multimodal Understanding (1/1)
On average across the 3 shared benchmarks, Gemini 3.1 Flash-Lite scores 2.93 higher.
Largest single-benchmark gap: SciCode — Gemini 3.1 Flash-Lite 43.40 vs Qwen3.6-35B-A3B 36.60 (+6.80).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.