Qwen 3.8 27b:thinking API

Alibaba's Qwen 3.8 27b:thinking, through an OpenAI-compatible API. 2 providers quote it now, and each request goes to the offer with the lowest estimated charge that meets your rules.

Providers compete for each request. The exchange ranks the offers that meet your rules by estimated charge. It fixes your maximum charge before the first token, and you pay for the tokens that the model uses.
The OpenAI and Anthropic APIs. Use the OpenAI or Anthropic SDK that you have. Set the base URL and your key, and send qwen3.8-27b:thinking as the model.
Your data rules. Send x-liquid-retention: none and only offers whose provider does not keep your content compete for the request.

Facts

Model name: qwen3.8-27b:thinking
Maker: Alibaba
Context window: 262.1K tokens
Input price: $0.074 to $0.100 per million tokens
Output price: $0.400 to $0.987 per million tokens
Providers: 2 independent providers
Regions: global, us

Providers

Fortytwo: us, $0.100 input and $0.400 output per million tokens, Free cache read per million tokens, 262.1K tokens of context
Munito AI: global, $0.074 input and $0.987 output per million tokens, $0.021 cache read per million tokens, 262.1K tokens of context

Scores

Intelligence: 20.1, #28 of 41
HLE: 33.9, #72 of 273
GPQA: 90.5, #41 of 250
AA-LCR: 82.0, #28 of 237
SciCode: 46.6, #74 of 123
SimpleQA Verified: 45.8, #40 of 126

Scores: Artificial Analysis (https://artificialanalysis.ai/)

Scores: Epoch AI, 'Capabilities & Benchmarking' (https://epoch.ai/benchmarks)

Request

curl https://router.inference.ai.exchange/v1/chat/completions \
  -H "Authorization: Bearer $LIQUID_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen3.8-27b:thinking",
    "messages": [{"role":"user","content":"Say hello."}]
  }'

Questions

What is Qwen 3.8 27b:thinking?: Qwen 3.8 27b:thinking is a model from Alibaba.
How much does Qwen 3.8 27b:thinking cost?: Offers now ask $0.074 to $0.100 per million input tokens and $0.400 to $0.987 per million output tokens. A job pays for the tokens it uses. The exchange picks the cheapest eligible offer.
What is the context length of Qwen 3.8 27b:thinking?: The largest offer takes 262,144 input tokens. One job can generate up to 235,929 tokens. A smaller offer can take less. The exchange skips an offer that cannot take your message.
What can Qwen 3.8 27b:thinking do?: Its providers list Vision, Reasoning effort, JSON object, JSON schema, Frequency penalty, Log probabilities, Presence penalty, Repetition penalty, Seed, Stop sequences, Top-k sampling, Top-p sampling, Text, Tools.
What inputs and outputs does Qwen 3.8 27b:thinking support?: Text · Images → Text.
Which providers serve Qwen 3.8 27b:thinking?: 2 independent providers quote it now, with 2 offers. The table above ranks them by the cost of a typical job.
How do I call Qwen 3.8 27b:thinking?: Send "qwen3.8-27b:thinking" as the model in an OpenAI or Anthropic request. The API section links to setup steps for each client.
When was Qwen 3.8 27b:thinking listed?: The exchange first listed it on 10/10/2026. This is not its release date.