Google: Gemma 4 31B

google/gemma-4-31b-it

Gemma 4 31B is a 31B open-weight model, ranking #91 of 235 on the TryAii Score. Its best showing is #18 on IFBench. It's served by 20 independent hosts from $0.09/$0.34 per million input/output tokens — about a quarter of what models at its level usually charge.

Prompt $/1M

$0.090

Completion $/1M

$0.340

Context

262,144 tok

Best provider

511 tok/s · Cerebras

Specs

Model id: google/gemma-4-31b-it
Display name: Google: Gemma 4 31B
Modality: text+image+video->text
Tokenizer: Gemma
Instruct type: —
Context length: 262,144
Max completion: 16,384
Prompt $/1M: $0.090
Completion $/1M: $0.340
Free tier: no

Capabilities

Function callingno
JSON modeno
System promptno
Visionyes
Moderatedno

Benchmark scores

Each axis is one benchmark, normalized to the best model in the field.

No benchmark data for this model.

Speed by provider

Provider	tok/s	TTFT	Total latency
Cerebrasfastest	511	283 ms	393 ms
ModelRun	119	86 ms	137 ms
Friendli	117	519 ms	777 ms
Artificial Analysis	73	589 ms	—
CoreWeave	50	767 ms	1,305 ms
SambaNova	42	1,478 ms	1,681 ms
Crusoe	39	1,070 ms	1,656 ms
OpenInference	38	434 ms	723 ms
WandB	35	444 ms	633 ms
Venice	34	1,270 ms	4,652 ms
SiliconFlow	30	14,011 ms	17,061 ms
Parasail	26	707 ms	1,077 ms
DeepInfra	24	2,585 ms	3,648 ms
Novita	21	1,080 ms	1,509 ms
Morph	16	4,713 ms	6,692 ms
Phala	14	3,731 ms	5,214 ms
Chutes	13	2,505 ms	4,728 ms
Together	10	486 ms	648 ms
Ambient	5	2,877 ms	4,510 ms
AkashML	4	4,466 ms	7,768 ms
NextBit	4	2,413 ms	3,085 ms
googlefastest	—	—	—