Qwen3.8-27B vs Qwen2.5-VL-3B-Instruct-AWQ
Specs, VRAM requirements and download trends — updated 2 November 2026.
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | Qwen3.8-27B | Qwen2.5-VL-3B-Instruct-AWQ |
|---|---|---|
| Deployment | Open weights · also via API | Open weights |
| Parameters | 27.8 B | 3.8 B |
| Architecture | Qwen3_5ForConditionalGeneration | Qwen2_5_VLForConditionalGeneration |
| Context length | 262,144 | 128,000 |
| Licence | apache-2.0 | — |
| Commercial use | Allowed | Unknown |
| Languages | — | 1 |
| Downloads 30d | 7,508,158 | 867,836 |
| Downloads all time | 22.0 M | 4.2 M |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2026-08-05 | 2025-02-13 |
| API input price / 1M tokens | $0.15 | not on OpenRouter |
| API output price / 1M tokens | $1.88 | not on OpenRouter |
| Est. cost per 1M input + 1M output tokens | $2.03 | not on OpenRouter |
| Cached input / 1M tokens | $0.038 | not on OpenRouter |
| AA Intelligence Index | 20.1 | not on OpenRouter |
| AA Coding Index | 68.1 | not on OpenRouter |
| AA Agentic Index | 45.8 | not on OpenRouter |
| Max output tokens | 235,929 | not on OpenRouter |
| API providers | 25 | not on OpenRouter |
| Tool calling | Yes | not on OpenRouter |
| Reasoning mode | Yes | not on OpenRouter |
API cost
Estimated cost of 1M input + 1M output tokens via OpenRouter
Qwen3.8-27B
$2.03
Download trend
Daily snapshots, last 83 days (20 Aug – 10 Nov)
8.0 M867 K
Qwen3.8-27B
Qwen2.5-VL-3B-Instruct-AWQ
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
Qwen2.5-VL-3B-Instruct-AWQ · fp16 @ 125K ctx
13.0 GB / 24 GB
Qwen3.8-27B · fp16
65.8 GB / 24 GB