---
id: "qwen/qwen3-8b"
also: ["qwen/qwen3-8b-04-28", "Qwen/Qwen3-8B", "hf/Qwen/Qwen3-8B"]
cheapest_list_usd_per_m: "0.07 in, 0.18 out, at nscale on hf"
cheapest_effective_usd_per_m: "0.07 in, 0.18 out, at nscale on hf"
context: 131072
max_output: 8192
input: ["text"]
output: ["text"]
released: "2025-04-28"
params: ["frequency_penalty", "include_reasoning", "max_tokens", "presence_penalty", "reasoning", "response_format", "seed", "stop", "temperature", "tool_choice", "tools", "top_k", "top_p"]
endpoints: 3
gateways: "hf (2), openrouter (1)"
no_training: 1
trains: 0
retains_prompts: 1
intelligence: 7.3000
coding: 9
agentic: 0.8000
---
# Qwen: Qwen3 8B

Qwen3-8B is a dense 8.2B parameter causal language model from the Qwen3 series, designed for both reasoning-heavy tasks and efficient dialogue. It supports seamless switching between "thinking" mode for math,...