Qwen3-VL-8B-Instruct vs Qwen3.8-27B-MLX-4bit
Specs, VRAM requirements and download trends — updated 10 November 2026.
Qwen · A · Open weights
Params
8.8 B
Context
—
30d
7.1 M
apache-2.0 · commercial OK
lmstudio-community · B · Open weights
Params
4.7 B
Context
—
30d
3.0 M
apache-2.0 · commercial OK
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | Qwen3-VL-8B-Instruct | Qwen3.8-27B-MLX-4bit |
|---|---|---|
| Deployment | Open weights · also via API | Open weights |
| Parameters | 8.8 B | 4.7 B |
| Architecture | Qwen3VLForConditionalGeneration | Qwen3_5ForConditionalGeneration |
| Context length | 262,144 | — |
| Licence | apache-2.0 | apache-2.0 |
| Commercial use | Allowed | Allowed |
| Languages | — | — |
| Downloads 30d | 7,092,774 | 3,048,666 |
| Downloads all time | 79.7 M | 11.7 M |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2025-10-11 | 2026-08-14 |
| API input price / 1M tokens | $0.25 | not on OpenRouter |
| API output price / 1M tokens | $0.75 | not on OpenRouter |
| Est. cost per 1M input + 1M output tokens | $1.00 | not on OpenRouter |
| Cached input / 1M tokens | $0.12 | not on OpenRouter |
| AA Intelligence Index | — | not on OpenRouter |
| AA Coding Index | — | not on OpenRouter |
| AA Agentic Index | — | not on OpenRouter |
| Max output tokens | 235,929 | not on OpenRouter |
| API providers | 1 | not on OpenRouter |
| Tool calling | Yes | not on OpenRouter |
| Reasoning mode | No | not on OpenRouter |
API cost
Estimated cost of 1M input + 1M output tokens via OpenRouter
Qwen3-VL-8B-Instruct
$1.00
Download trend
Daily snapshots, last 90 days (13 Aug – 10 Nov)
20.2 M890 K
Qwen3-VL-8B-Instruct
Qwen3.8-27B-MLX-4bit
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
Qwen3.8-27B-MLX-4bit · fp16
18.9 GB / 24 GB
Qwen3-VL-8B-Instruct · fp16
21.1 GB / 24 GB