Qwen3.6-35B-A3BvsQwen3.5-35B-A3B
Across 3 shared benchmarks, Qwen3.6-35B-A3B leads overall: Qwen3.6-35B-A3B wins 3, Qwen3.5-35B-A3B wins 0, with 0 ties and an average score difference of +78.13.
Qwen3.6-35B-A3B
阿里巴巴 · 2026-04-16 · Reasoning model
Qwen3.5-35B-A3B
阿里巴巴 · 2026-02-25 · Reasoning model
Qwen3.6-35B-A3B3 wins(100%)(0%)0 winsQwen3.5-35B-A3B
Benchmark scores
Grouped by capability, sorted by largest gap within each. 3 shared benchmarks.
Long Context
Qwen3.6-35B-A3B 1/1| Benchmark | Qwen3.6-35B-A3B | Qwen3.5-35B-A3B | Diff |
|---|---|---|---|
| AA-LCR | 64.30120 / 170Normal (No Tools) | 63124 / 170Normal (No Tools) | +1.30 |
Productivity Knowledge
Qwen3.6-35B-A3B 1/1| Benchmark | Qwen3.6-35B-A3B | Qwen3.5-35B-A3B | Diff |
|---|---|---|---|
| GDPval-AA v2 | 95391 / 105Normal (With Tools) | 740100 / 105Normal (With Tools) | +213 |
Text Embedding
Qwen3.6-35B-A3B 1/1| Benchmark | Qwen3.6-35B-A3B | Qwen3.5-35B-A3B | Diff |
|---|---|---|---|
| Context Arena | 53.2084 / 126Normal (No Tools) | 33.10111 / 126Normal (No Tools) | +20.10 |
Specs
| Field | Qwen3.6-35B-A3B | Qwen3.5-35B-A3B |
|---|---|---|
| Publisher | 阿里巴巴 | 阿里巴巴 |
| Release date | 2026-04-16 | 2026-02-25 |
| Model type | Reasoning model | Reasoning model |
| Architecture | MoE | MoE |
| Parameters | 35B | 35B |
| Context length | 200K | 262K |
| Max output | 80K | Not available |
Summary
- Qwen3.6-35B-A3Bleads in:Long Context (1/1), Productivity Knowledge (1/1), Text Embedding (1/1)
On average across the 3 shared benchmarks, Qwen3.6-35B-A3B scores 78.13 higher.
Largest single-benchmark gap: GDPval-AA v2 — Qwen3.6-35B-A3B 953 vs Qwen3.5-35B-A3B 740 (+213).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.