Gemma 3 27B API pricing

7 providers serve Gemma 3 27B. Gemma 3 27B is an open-weight model (Gemma) with 27B total parameters, up to 262K 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 $0.060/1M at Nebius Output floor $0.160/1M at DeepInfra Find your workload winner →
Provider Input $/1M Output $/1M Cache read $/1M Context
Nebius $0.060 $0.200 - 128K Use
OpenRouter marketplace quote $0.080 $0.450 $0.040 262K Use
DeepInfra $0.090 $0.160output floor - 131K Use
Novita AI $0.119 $0.200 - 98K Use
AWS Bedrock $0.230 $0.380 - 128K Use
Scaleway $0.250 $0.500 - 40K Use
Fireworks AI $0.900 $0.900 - 131K Use

The cheapest way to run Gemma 3 27B

On an illustrative 70M-input / 30M-output monthly workload, today's lowest listed cost is roughly $10.20/month, at Nebius. 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, the API floor remains below our self-hosting estimate of ~$0.239/1M. Details 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 27B self-hosting economics

Gemma 3 27B is open-weight (Gemma), so the API price above competes with the GPU-hour market. 27B total parameters needs roughly 38 GB VRAM at FP8 or 19 GB at INT4, KV-cache headroom included.

GPU that fits (FP8, single card)VRAMFrom $/hrCheapest at
A6000 48 GB $0.287marketplace Vast.ai Rent
A40 48 GB $0.350community RunPod Rent
6000 Ada 48 GB $0.388marketplace Vast.ai Rent
A100 SXM 40GB 40 GB $0.388marketplace Vast.ai Rent
L40 48 GB $0.402marketplace Vast.ai Rent
A100 PCIe 80 GB $0.442marketplace Vast.ai Rent
Breakeven estimate: a well-batched vLLM deployment on one A6000 ($0.287/hr) at ~667 aggregate tok/s and 50% utilization works out to roughly $0.239 per 1M output tokens, versus $0.160 via the cheapest output-token API (DeepInfra). The API is cheaper unless your utilization or throughput beats these assumptions. 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 12B $0.050 $0.100 5
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 27B?

As of 2026-08-23, the lowest input price for Gemma 3 27B is Nebius at $0.060 per 1M input tokens. The lowest output price is DeepInfra at $0.160 per 1M output tokens. The cheapest provider for a real workload depends on its input/output mix. The next-lowest input price is OpenRouter at $0.080, a 33% difference.

How much VRAM do you need to self-host Gemma 3 27B?

Gemma 3 27B has 27B parameters, so plan for roughly 38 GB of VRAM at FP8 or 19 GB at INT4/AWQ, KV-cache headroom included. A single A6000 (48 GB) fits it.

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