Models · Compare
Qwen3.5 35B-A3B vs Claude Sonnet 4.6
Rows highlighted in warm gray are where the models differ. Numbers carry their as-of date and primary source.
Specs
| Field | A: Qwen3.5 35B-A3B | B: Claude Sonnet 4.6 |
|---|---|---|
| Released | 2026-02-16 | 2026-02-17 |
| Developer | Alibaba | Anthropic |
| Openness | Open | Proprietary |
| License | Apache-2.0 | Proprietary |
| OSI-approved | yes | no |
| Data released | no | no |
| Training code | no | no |
| Architecture | moe | unknown |
| Total params | 35B | — |
| Active params | 3B | — |
| Experts | — | — |
| Context window | 262K | 1.0M |
| Attention | gqa | unknown |
| Position enc. | rope-yarn | unknown |
| Pretraining tokens | — | — |
| Post-training | sft, rlhf | rlhf, constitutional |
| Training hardware | — | — |
| $/M input | — | $3.00 |
| $/M output | — | $15.00 |
| Output tok/sec | — | 49 |
Benchmarks
Missing scores render as not reported; never inferred. Bold highlights the leader per benchmark.
General reasoning
| GPQA-Diamond | — | 79.9 2026-05-21 |
Code
| SWE-Bench Verified | — | 80.2 2026-02-17 |
Context · A
A small Qwen3.5 mixture-of-experts: about 35B total with roughly 3B active per token, so it decodes very fast for its size. A pure-dense 35B at Q8 would be memory-bandwidth-bound near single-digit tokens/sec on a 256 GB/s box; the measured high speed on Strix Halo is itself evidence this SKU is sparse, not dense. It is typically run at Q8.
Context · B
Mid-tier Sonnet refresh shipped February 17 2026 at the same $3 / $15 pricing as Sonnet 4.5, with a 1M context window in beta. Default model for Free and Pro plans on Claude.ai at launch. Anthropic reported Claude Code users preferred Sonnet 4.6 over Sonnet 4.5 roughly 70 percent of the time.