gemma-4-31B-it-FP8-block vs SmolVLM-256M-Instruct
Specs, VRAM requirements and download trends — updated 28 July 2026.
RedHatAI · A · Open weights
Params
31.3 B
Context
—
30d
936 K
apache-2.0 · commercial OK
HuggingFaceTB · B · Open weights
Params
260 M
Context
—
30d
872 K
apache-2.0 · commercial OK
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | gemma-4-31B-it-FP8-block | SmolVLM-256M-Instruct |
|---|---|---|
| Parameters | 31.3 B | 260 M |
| Architecture | Gemma4ForConditionalGeneration | Idefics3ForConditionalGeneration |
| Context length | — | — |
| Licence | apache-2.0 | apache-2.0 |
| Commercial use | Allowed | Allowed |
| Languages | — | 1 |
| Downloads 30d | 936,323 | 871,636 |
| Downloads all time | 10.0 M | 7.5 M |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2026-04-03 | 2025-01-17 |
Download trend
Daily snapshots, last 90 days (13 Aug – 10 Nov)
3.4 M872 K
gemma-4-31B-it-FP8-block
SmolVLM-256M-Instruct
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
SmolVLM-256M-Instruct · fp16
1.1 GB / 24 GB
gemma-4-31B-it-FP8-block · fp16
41.8 GB / 24 GB