Gemma 3 12B API pricing

5 providers serve Gemma 3 12B. Gemma 3 12B is an open-weight model (Gemma) with 12B total parameters, up to 131K context. Prices below are per 1M tokens, cheapest input first (self-hosting math further down).

Refreshed 2026-08-23 · list prices from provider APIs
Input floor $0.050/1M at DeepInfra, Novita AI and OpenRouter Output floor $0.100/1M at DeepInfra, Novita AI and Crusoe Find your workload winner →
Provider Input $/1M Output $/1M Cache read $/1M Context
DeepInfra $0.050 $0.100output floor - 131K Use
Novita AI $0.050 $0.100 - 131K Use
OpenRouter marketplace quote $0.050 $0.150 - 131K Use
AWS Bedrock $0.090 $0.290 - 128K Use
Crusoe $0.100 $0.100 - 131K Use

The cheapest way to run Gemma 3 12B

On an illustrative 70M-input / 30M-output monthly workload, today's lowest listed cost is roughly $6.50/month — the same at DeepInfra and Novita AI. Input-heavy, output-heavy and cache-heavy workloads can produce different winners. Use the live mix calculator below rather than combining floors from two different providers. On output-token cost alone, a busy self-hosted deployment can work out cheaper (~$0.022/1M output on V100). See the math below. Run your own numbers in the breakeven calculator, or read the breakeven math.

Your actual monthly cost

Uses today's list prices

Cache read uses the listed cache rate where available; otherwise it falls back to normal input price. Batch, write-cache, volume and negotiated discounts are excluded.

Or run it yourself: Gemma 3 12B self-hosting economics

Gemma 3 12B is open-weight (Gemma), so the API price above competes with the GPU-hour market. 12B total parameters needs roughly 17 GB VRAM at FP8 or 9 GB at INT4, KV-cache headroom included.

GPU that fits (FP8, single card)VRAMFrom $/hrCheapest at
V100 32 GB $0.022marketplace Vast.ai Rent
RTX 3090 24 GB $0.068marketplace Vast.ai Rent
RTX 4090 24 GB $0.134marketplace Vast.ai Rent
RTX A5000 24 GB $0.160community RunPod Rent
RTX 4000 SFF Ada Generation 20 GB $0.180community RunPod Rent
RTX A4500 20 GB $0.190community RunPod Rent
Breakeven estimate: a well-batched vLLM deployment on one V100 ($0.022/hr) at ~563 aggregate tok/s and 50% utilization works out to roughly $0.022 per 1M output tokens, versus $0.100 via the cheapest output-token API (DeepInfra, Novita AI and Crusoe). Self-hosting wins on cost if you can keep the GPU busy. Planning estimate: throughput varies with hardware, quantization, batch size and context. Tune it in the calculator.

Related models

ModelCheapest in $/1MCheapest out $/1MProviders
Gemma 3 27B $0.060 $0.160 7
Gemma 4 26B A4B $0.070 $0.300 4
Gemma 4 31B $0.100 $0.340 4
Gemma-7b-It $0.050 $0.080 3
Gemma 3 4B $0.040 $0.080 3

FAQ

What is the cheapest API for Gemma 3 12B?

As of 2026-08-23, the lowest input price for Gemma 3 12B, $0.050 per 1M tokens, is the same list price at DeepInfra, Novita AI and OpenRouter. The lowest output price, $0.100 per 1M tokens, is likewise shared by DeepInfra, Novita AI and Crusoe. The cheapest provider for a real workload depends on its input/output mix. The next-lowest input price is AWS Bedrock at $0.090, a 80% difference.

How much VRAM do you need to self-host Gemma 3 12B?

Gemma 3 12B has 12B parameters, so plan for roughly 17 GB of VRAM at FP8 or 9 GB at INT4/AWQ, KV-cache headroom included. A single V100 (32 GB) fits it.

← All models · GPU rental prices · Breakeven calculator · Get this data as JSON