Home / Models / Qwen (Alibaba) / Qwen3 8B

Qwen3 8B pricing and routing profile

Qwen3 8B from Qwen (Alibaba) lists at $0.117 per million input tokens and $0.455 per million output tokens, with a 131,072-token context window. Prices from OpenRouter's catalogue on 2026-09-28.

Input, per 1M tokens$0.117
Output, per 1M tokens$0.455
Context window131K
Confidence signalNot exposed

Tool callingReasoning control

Cost per 1,000 requests

Request shapeList cost
Classification or routing (600 tokens in, 10 out)$0.0747
Chat reply (2,000 in, 400 out)$0.416
Retrieval-augmented answer (8,000 in, 500 out)$1.16

Before prompt caching. Try your own shape in the cost calculator.

Can Gimbix route to it first?

Not as a first-answer model through OpenRouter: no host returns token log probabilities for Qwen3 8B. It can still be the model Gimbix falls back to when a cheaper model is unsure.

Cheaper models that can answer first

Models priced below Qwen3 8B that expose a confidence signal. Whether they answer your traffic as well is what a Gimbix pilot measures.

ModelInput / output per 1MCheaper on a chat reply
Qwen3 VL 32B Instruct$0.104 / $0.41610%
Nemotron 3 Super$0.08 / $0.4518%
Gemma 3 27B$0.08 / $0.4518%
Llama 3.3 70B Instruct$0.10 / $0.3221%
Qwen3 30B A3B Instruct 2507$0.10 / $0.3023%
Gemma 4 31B$0.09 / $0.3424%

Questions

How much does Qwen3 8B cost?

Qwen3 8B lists at $0.117 per million input tokens and $0.455 per million output tokens on OpenRouter as of 2026-09-28. A typical chat reply of 2,000 input and 400 output tokens costs about $0.416 per 1,000 requests.

What is the context window of Qwen3 8B?

131,072 tokens.

Can a cheaper model replace Qwen3 8B?

Sometimes. Gimbix proves it on your own traffic before routing: the cheaper model must match Qwen3 8B on one half of your requests and make at most three answers worse on the other.