Qwen2-1.5B-Instruct vs PowerMoE-3b
Specs, VRAM requirements and download trends — updated 14 August 2026.
Qwen · A · Open weights
Params
1.5 B
Context
32,768
30d
210 K
apache-2.0 · commercial OK
ibm-research · B · Open weights
Params
3.4 B
Context
4,096
30d
171 K
apache-2.0 · commercial OK
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | Qwen2-1.5B-Instruct | PowerMoE-3b |
|---|---|---|
| Parameters | 1.5 B | 3.4 B |
| Architecture | Qwen2ForCausalLM | GraniteMoeForCausalLM |
| Context length | 32,768 | 4,096 |
| Licence | apache-2.0 | apache-2.0 |
| Commercial use | Allowed | Allowed |
| Languages | 1 | — |
| Downloads 30d | 210,398 | 170,877 |
| Downloads all time | 31.9 M | 12.0 M |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2024-06-03 | 2024-08-14 |
Download trend
Daily snapshots, last 90 days (28 Jul – 25 Oct)
3.3 M171 K
Qwen2-1.5B-Instruct
PowerMoE-3b
A overtook B on 2 Oct
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
Qwen2-1.5B-Instruct · fp16
4.1 GB / 24 GB
PowerMoE-3b · fp16
15.9 GB / 24 GB