Qwen

Qwen3 8B

CTX131K

50Quality

Reasoning · Tools

Input$0.18/M

Output$0.70/M

Qwen instruction model for multilingual chat, reasoning, and tool use

Open weightsApache 2.0526/1040

Drag to turn

QwenTextCommon

Qwen3 8B

Qwen instruction model for multilingual chat, reasoning, and tool use

qwen3-8b

Type: Text
Quality: 50 · Top 50% on the Epoch Capabilities Index
Context window: 131,072 tokens
Max output: 8,192 tokens
Input price: $0.18 / 1M tokens
Output price: $0.70 / 1M tokens
Cache read: —
Cache write: —
Parameters: 8.2B (8,190,735,360)
Downloads: 9.1M in the last 30 days
Likes: 2.1K
Access to weights: Open download
Released: Apr 1, 2025
Knowledge cutoff: Apr 2025
Takes in: Text
Puts out: Text
Reasoning: Yes
Tool calling: Yes
Structured output: No
File attachments: No
Access: Open weights
License: Apache 2.0
Rarity: Common
Set number: 526/1040

Open Qwen docs Weights on Hugging Face Quality 50 Where to run it 8 Evolution 5/8 Race it 134 tok/s Meet its creature Battle Mercury Edit 2

License read from the model’s Hugging Face card. Specs from models.dev, downloads from Hugging Face, last changed Oct 18, 2026. Official source: www.alibabacloud.com.

How good is it?

Public leaderboard scores for Qwen3 8B, refreshed daily from Epoch AI and LMArena.

Top 50% of the 277 models on the Epoch Capabilities Index: #139.

Quality is the share of a leaderboard’s models this one beats, from 0 to 100. Chat models are rated on Epoch AI’s Capabilities Index when it lists them, otherwise on LMArena; image and video models on LMArena’s image and video arenas.

Leaderboards

Epoch Capabilities Index#139 of 277
One number fitted from dozens of hard benchmarks, from science questions to agentic coding.136.2 index

Benchmarks Epoch AI runs

Scores from Epoch AI, licensed CC BY 4.0. Boards list models under their own names, so a close match is shown with the name it was scored under. Credits

Where to run it

6 providers sell Qwen3 8B through an API, and 2 Hugging Face hosts serve it. Prices are per million tokens, cheapest first.

CheapestSiliconFlow$0.06 in · $0.06 out
From the labAlibaba$0.18 in · $0.70 out
Fastest on Hugging FaceNscale134 tokens/s

API providers

Provider	Input	Output	Context
SiliconFlowCheapest paid`Qwen/Qwen3-8B`	$0.06	$0.06	131K
SiliconFlow (China)`Qwen/Qwen3-8B`	$0.06	$0.06	131K
Pioneer`Qwen/Qwen3-8B`	$0.20	$0.20	41K
Alibaba (China)`qwen3-8b`	$0.072	$0.29	131K
NanoGPT`qwen/qwen3-8b`	$0.47	$0.47	41K
AlibabaOfficial`qwen3-8b`	$0.18	$0.70	131K

Hugging Face Inference Providers

Host	Input	Output	Context	Speed	First token
NscaleFastestTools	$0.07	$0.18	41K	134 tok/s	0.6s
Featherless	—	—	—	—	—

From models.dev and Hugging Face, checked daily. Resellers’ listings can lag behind their price pages, so confirm with the provider before you commit.

Face off

Current text models from other labs at a similar output price.

Qwen3 8B

Qwen3 8B

How good is it?

Leaderboards

Benchmarks Epoch AI runs

Where to run it

API providers

Hugging Face Inference Providers

Face off

More from Qwen

Qwen-Image-2.1-Turbo

Qwen3.8 Omni Flash

Qwen-Image-2.1

Qwen3.8 Flash

Qwen3.8-Flash-Next

Qwen3.8-2.4T-A95B

Qwen3.8-27B

Qwen3.8 Max