Google Gemini and Gemma · LLM API pricing

Gemma 4 31B IT API pricing: cheapest provider

6 providers sell Gemma 4 31B IT. W&B Inference (CoreWeave) is cheapest; the most expensive charges 5.1× as much for the same model.

Cheapest$0.16per 1M · W&B Inference (CoreWeave)
Input · output$0.10 / $0.34per 1M tokens
Price gap5.1×across 6 providers
Accuracy75.8%GPQA · minimal reasoning
Context262Kmax tokens

$0.16 W&B Inference (CoreWeave)5.1× spread$0.8125 SiliconFlow

Compare all 6 providers ⇄ Compare with another model

Already paying for Gemma 4 31B IT? Check if you overpay →

Standard list prices in USD from each provider's pricing page · last price change Oct 28, 2026 · How we collect and rank prices

Ranked for your volume

Gemma 4 31B IT monthly cost by provider

Change the monthly tokens and the ranking updates instantly, using each provider's standard list price.

Input tokens per month (millions)Prompts, context and documents you send.Output tokens per month (millions)Tokens the model generates.

More requirements: caching, context, features, provider

Cached share of input (%)0 = no caching. Only documented cache-read rates qualify.Required context per request (tokens)0 = any. Providers without a sufficient known limit are excluded.Required featureProvider requirement

Provider	Estimated / month	Input / 1M	Output / 1M	Cached input / 1M	Context
1 W&B Inference (CoreWeave)Inference provider - lowest for your volumePricing page	$1.68	$0.10	$0.34	Not listed	262,000
2 DeepInfraInference providerPricing page	$2.06	$0.13	$0.38	Not listed	262,144
3 FriendliAIInference providerPricing page	$2.20	$0.14	$0.40	Not listed	262,144
4 Novita AIInference providerPricing page	$2.20	$0.14	$0.40	Not listed	262,144
5 SambaNovaInference providerPricing page	$6.10	$0.38	$1.15	Not listed	131,072
6 SiliconFlowInference providerPricing page	$9.50	$0.75	$1.00	$0.25	262,144

Order depends on listed price for your volume only. No provider pays for its position.

Input volume includes the cached share, not an additional token volume. This is a list-price estimate, not a quote: cache-write charges, cache eligibility/TTL, batch discounts, long-context surcharges, regional pricing and taxes are excluded. Check the provider's terms for your workload before switching.

Share this comparison as a Markdown table

For a post, a doc or a forum answer. The source line links back here, so readers can check today's prices.

Markdown

| Gemma 4 31B IT provider | Input / 1M tokens | Output / 1M tokens | 10M in + 2M out per month |
|---|---:|---:|---:|
| W&B Inference (CoreWeave) | $0.10 | $0.34 | $1.68 |
| DeepInfra | $0.13 | $0.38 | $2.06 |
| FriendliAI | $0.14 | $0.40 | $2.20 |
| Novita AI | $0.14 | $0.40 | $2.20 |
| SambaNova | $0.38 | $1.15 | $6.10 |
| SiliconFlow | $0.75 | $1.00 | $9.50 |

Standard list prices as of Oct 31, 2026. Source: [AISelect.online](https://aiselect.online/models/gemma-4-31b-it)

Accuracy: GPQA Diamond (graduate-level science questions) measured by Epoch AI (CC BY 4.0) on gemma-4-31b-it_minimal, run Aug 6, 2026. The model is the same at every provider, though quantized deployments can score lower.

Price history

The cheapest price has not changed since tracking started on Sep 28, 2026: $0.16 per 1M tokens at W&B Inference (CoreWeave). All price changes

Switch to the cheapest provider

Switch to W&B Inference (CoreWeave)

Already using the OpenAI SDK? Change the base URL, key and model name:

from openai import OpenAI
import os

client = OpenAI(
    base_url="https://api.inference.wandb.ai/v1",
    api_key=os.environ["WANDB_API_KEY"],
)
response = client.chat.completions.create(
    model="google/gemma-4-31B-it",
    messages=[{"role": "user", "content": "Hello"}],
)

Base URL from W&B Inference (CoreWeave)'s API docs, model name from the open pricing datasets. Check the model page in W&B Inference (CoreWeave)'s docs before you switch production traffic.

Compare Gemma 4 31B IT with another model

Or pick any model

Model A

Model B

Before you pick a provider

Context windows differ by provider, from 131K to 262K tokens. A cheaper host may not accept your longest prompts.

1 of 6 providers list a cached-input price. Repeated system prompts or documents can cost much less there than the standard input rate shown above.

Keep comparing

More Google Gemini and Gemma models

Gemma 4 26B A4B IT, Gemma-3-27b-it, Gemini 2.5 Flash, Gemini 2.5 Pro, Gemini 3.1 Flash Lite, Gemini 3.5 Flash, Gemini 3.7 Flash, Gemma 3 12B IT

All Google Gemini and Gemma API prices - All LLM API prices - Watch Gemma 4 31B IT prices (RSS) - Methodology