gemma-4-31B-it vs gemma-3-27b-it

Specs, VRAM requirements and download trends — updated 31 July 2026.

google · A · Open weights
Params
32.7 B
Context
—
30d
7.1 M
apache-2.0 · commercial OK
google · B · Open weights
Params
27.4 B
Context
—
30d
857 K
gemma licence · conditional

Specification comparison

Differences are highlighted; identical values are muted.

Specification gemma-4-31B-it gemma-3-27b-it
Deployment Open weights · also via API Open weights · also via API
Parameters 32.7 B 27.4 B
Architecture Gemma4ForConditionalGeneration Gemma3ForConditionalGeneration
Context length 262,144 131,072
Licence apache-2.0 gemma
Commercial use Allowed Conditional
Languages — —
Downloads 30d 7,079,469 857,335
Downloads all time 73.4 M 16.3 M
Quantizations on the Hub SAFETENSORS SAFETENSORS
Gated No Yes — accept licence
First seen on the Hub 2026-03-11 2025-03-01
API input price / 1M tokens $0.20 $0.08
API output price / 1M tokens $0.40 $0.45
Est. cost per 1M input + 1M output tokens $0.60 $0.53
Cached input / 1M tokens — $0.04
AA Intelligence Index 4.4 —
AA Coding Index 43.4 10.1
AA Agentic Index 4.2 0.1
Max output tokens 16,384 117,964
API providers 16 4
Tool calling Yes Yes
Reasoning mode Yes No

API cost

Estimated cost of 1M input + 1M output tokens via OpenRouter

gemma-4-31B-it $0.60
gemma-3-27b-it $0.53

gemma-3-27b-it is 1.1× cheaper at list price.

Download trend

Daily snapshots, last 90 days (13 Aug – 10 Nov)

10.0 M857 K
gemma-4-31B-it gemma-3-27b-it

VRAM side by side

On RTX 4090 · 24 GB · 8K context unless noted

gemma-3-27b-it · fp16 65.0 GB / 24 GB
❌ Won’t fit
gemma-4-31B-it · fp16 74.2 GB / 24 GB
❌ Won’t fit

Adjacent comparisons