Last refreshed 2026-10-26. Next refresh: weekly.
Why use Gemma 3 12B on Fireworks AI?
Fireworks AI offers Gemma 3 12B with pay-as-you-go pricing at $0.20/1M input tokens. Fireworks AI offers a generative AI platform as a service, focusing on rapid product iteration and cost-efficient AI deployment.
Compare Gemma 3 12B across 6 providers to find the best fit for your use caseInput / 1M
$0.20
Output / 1M
$0.20
Cache
Not sourced
Batch
Not sourced
Setup recipe
Python + curlInstall
pip install openaiAuth
export FIREWORKS_API_KEY=...Call
import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["FIREWORKS_API_KEY"],Model ID
accounts/fireworks/models/gemma-3-12b-itRequest example
import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["FIREWORKS_API_KEY"],
base_url="https://api.fireworks.ai/inference/v1"
)
response = client.chat.completions.create(
model="accounts/fireworks/models/gemma-3-12b-it",
messages=[{"role": "user", "content": "Hello"}]
)
print(response.choices[0].message.content)Gotchas
- Use provider model ID "accounts/fireworks/models/gemma-3-12b-it", not the LLMReference slug "gemma-3-12b-it".
- Fireworks model IDs use "accounts/fireworks/models/{model-name}" format, e.g. "accounts/fireworks/models/llama4-scout-instruct-basic" or "accounts/fireworks/models/deepseek-r1".
- The examples expect FIREWORKS_API_KEY; rename it only if your application config maps the new variable.
Compare Gemma 3 12B Across Providers
| Provider | Input (per 1M) | Output (per 1M) |
|---|---|---|
| Cloudflare Workers AI | $0.34 | $0.56 |
| Fireworks AI | $0.20 | $0.20 |
| AWS Bedrock | $0.30 | $0.30 |
| OpenRouter | $0.05 | $0.15 |
| GCP Vertex AI | $0.04 | $0.13 |
Pricing
| Type | Price (per 1M) |
|---|---|
| Input tokens | $0.20 |
| Output tokens | $0.20 |
Capabilities
Structured Outputs
About Gemma 3 12B
Gemma 3 12B is Google DeepMind's Gemma 3 model. It offers a 128K-token context window with weights openly available for self-hosting.
Get Started
Model Specs
Released2025-03-12
Parameters12B
Context128K
ArchitectureDecoder Only
Knowledge cutoff2024-08