Qwen2.5-32B-Instruct vs Qwen3.6-35B-A3B-abliterated-v4
Specs, VRAM requirements and download trends — updated 21 September 2026.
Bahushruth · B
Params
34.7 B
Context
262,144
30d
877 K
apache-2.0 · commercial OK
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | Qwen2.5-32B-Instruct | Qwen3.6-35B-A3B-abliterated-v4 |
|---|---|---|
| Parameters | 32.8 B | 34.7 B |
| Architecture | Qwen2ForCausalLM | Qwen3_5MoeForCausalLM |
| Context length | 32,768 | 262,144 |
| Licence | apache-2.0 | apache-2.0 |
| Commercial use | Allowed | Allowed |
| Languages | 1 | — |
| Downloads 30d | 2,250,912 | 877,278 |
| Downloads all time | 30.9 M | 2.4 M |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2024-09-17 | 2026-06-11 |
Download trend
Daily snapshots, last 56 days (28 Jul – 21 Sep)
2.3 M630 K
Qwen2.5-32B-Instruct
Qwen3.6-35B-A3B-abliterated-v4
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
Qwen2.5-32B-Instruct · fp16
77.5 GB / 24 GB
Qwen3.6-35B-A3B-abliterated-v4 · fp16 @ 256K ctx
243.1 GB / 24 GB