Cerebras

Data handling can vary by model and deployment. The privacy badge above shows the more conservative of the provider-wide fallback and the policies for routes on this page; each model page shows the policy for that specific route.

Models served by Cerebras

Model name

Model	Context	Input /M	Output /M	Cache read /M	TTFT	Throughput
Gemma 4 31B FP16 google/gemma-4-31b-it:cerebras	262k	$1.04/M	$1.56/M	N/A	1.5s	1851 tps
Gemma 4 31B (Thinking) FP16 google/gemma-4-31b-it:thinking:cerebras	262k	$1.04/M	$1.56/M	N/A	1.5s	1851 tps
GPT OSS 120B FP16 openai/gpt-oss-120b:cerebras	128k	$0.37/M	$0.79/M	Supported	0.3s	619 tps
Qwen3.8 27B FP16 qwen3.8-27b:cerebras	131k	$1.04/M	$1.56/M	Supported	0.3s	374 tps
Qwen3.8 27B Thinking FP16 qwen3.8-27b:thinking:cerebras	131k	$1.04/M	$1.56/M	Supported	0.3s	374 tps

Gemma 4 31B

FP16

google/gemma-4-31b-it:cerebras

Context

262k

Input /M

$1.04/M

Output /M

$1.56/M

Cache read /M

N/A

TTFT

1.5s

Throughput

1851 tps

Gemma 4 31B (Thinking)

FP16

google/gemma-4-31b-it:thinking:cerebras

Context

262k

Input /M

$1.04/M

Output /M

$1.56/M

Cache read /M

N/A

TTFT

1.5s

Throughput

1851 tps

GPT OSS 120B

FP16

openai/gpt-oss-120b:cerebras

Context

128k

Input /M

$0.37/M

Output /M

$0.79/M

Cache read /M

Supported

TTFT

0.3s

Throughput

619 tps

Qwen3.8 27B

FP16

qwen3.8-27b:cerebras

Context

131k

Input /M

$1.04/M

Output /M

$1.56/M

Cache read /M

Supported

TTFT

0.3s

Throughput

374 tps

Qwen3.8 27B Thinking

FP16

qwen3.8-27b:thinking:cerebras

Context

131k

Input /M

$1.04/M

Output /M

$1.56/M

Cache read /M

Supported

TTFT

0.3s

Throughput

374 tps

Prices are per million tokens. TTFT is the observed time to first token. Selectable routes include the applicable provider-selection markup; fixed routes show the standard model price. Availability and pricing refresh continuously; each model page shows the live provider comparison.

Using Cerebras via the API

Call a model served by Cerebras using its displayed model ID, including the provider suffix when selection is supported.

curl https://nano-gpt.com/api/v1/chat/completions \
  -H "Authorization: Bearer $NANOGPT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "google/gemma-4-31b-it:cerebras",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

For models with provider selection, append :cerebras to the model ID, set "provider": "cerebras" in the request body, or send an X-Provider: cerebras header. Explicit provider selection adds a route-specific markup over that provider's base price; the prices in the table above already include it. Models without provider selection use the displayed model ID without a provider suffix or selection surcharge.

See the API documentation for provider preferences, price-aware routing, and error behavior.