All models

Qwen3.8 27B (Cerebras)

cerebras/qwen-3.8-27b

Pricing

Input
$0.400 / M
Output
$0.800 / M
Cached input
Cache write
Cache write (1h)
Currency
USD

Cents per million tokens. Every buffered reply carries X-Nozzle-Cost-Micro-Cents; streaming replies reconcile by request id.

Facts

Model id
cerebras/qwen-3.8-27b
Family
qwen3.8
Modality
text
Provider
cerebras
Upstream id
qwen-3.8-27b
Context window
131k
Max output tokens
131,072
Parameters
27B
License
Wire
openai_chat

Capabilities the router will accept

toolsparallel_tool_callstool_choice_namedjson_objectjson_schemareasoningtimestampstranslationstreamingseedpenalties

Run it on your own GPU

Blueprints: qwen38_sglang_bf16qwen38_sglang_fp8 · from $3.59 / GPU-hour

Status: callable