Qwen

Qwen 3 Coder — Benchmark Scores, Pricing & Performance Analysis

OPEN SOURCEQwen
Chatbot Arena ELO
1290
Output Speed
80 tok/s
Input Cost
$0.30/1M
Output Cost
$1.0/1M
Context Window
262K
Max Output
66K
Knowledge Cutoff
Jun 2025
Accepts
text

Qwen 3 Coder by Qwen demonstrates solid coding performance, competitive pricing. View detailed benchmark data including scores across coding, math, reasoning, speed, and cost metrics.

Specifications last checked against the provider on . Benchmark scores carry their own source, linked beside each number.

General Benchmarks

Chatbot Arena ELO
129047th of 74

Human preference ELO from blind head-to-head votes

LMSYS / HuggingFace
MMLU-Pro
72.0%42nd of 66

Massive Multitask Language Understanding professional benchmark

HuggingFace Leaderboard

Coding Benchmarks

LiveCodeBench
52.0%30th of 66

Live competitive programming benchmark

livecodebench.github.io
HumanEval+
90.0%11th of 66

Code generation correctness with extended tests

Papers
SWE-bench Verified
66.5%20th of 67

Real-world software engineering task resolution

swebench.com
Design Arena ELO
113574th of 120

Human preference ELO for building things — websites, UI, games, charts, SVG

Design Arena (via OpenRouter)

Math Benchmarks

GSM8K
90.0%42nd of 66

Grade school math word problems

Papers
MATH
76.0%40th of 66

Competition mathematics problem solving

Papers

Reasoning Benchmarks

ARC-AGI
28.0%32nd of 66

Abstraction and Reasoning Corpus for general intelligence

arcprize.org
GPQA Diamond
48.0%117th of 140

Graduate-level science Q&A by domain experts

Papers

Speed Benchmarks

Time to First Token
350 ms31st of 92

Latency before first token arrives

Aggregated
Output Speed
80 tok/s45th of 94

Tokens generated per second

Aggregated

Cost Benchmarks

Input Cost
$0.30204th of 413

Cost per 1M input tokens

OpenRouter API
Cached Input Cost
$0.1062nd of 136

Cost per 1M cached input tokens — the price that actually applies to a long agent conversation

OpenRouter API
Output Cost
$1.0197th of 413

Cost per 1M output tokens

OpenRouter API

Context Benchmarks

Context Length
262K108th of 451

Maximum context window size

OpenRouter API

Qwen 3 Coder — Benchmark Scores Overview

Scores normalized to percentage scale for visual comparison. ELO scores mapped to 0-100 range (1100-1500).

Qwen 3 Coder — Frequently Asked Questions

How intelligent is Qwen 3 Coder?

Qwen 3 Coder scores 1290 on the Chatbot Arena ELO rating, making it a mid-tier AI model. This score is based on blind head-to-head human preference voting.

How much does Qwen 3 Coder cost?

Qwen 3 Coder costs $0.30 per 1M input tokens and $1.0 per 1M output tokens. This makes it one of the more affordable models.

How fast is Qwen 3 Coder?

Qwen 3 Coder generates output at 80 tokens per second, which is moderate compared to other models. The time to first token is 350 ms.

How good is Qwen 3 Coder at coding?

Qwen 3 Coder achieves 66.5% on SWE-bench Verified, demonstrating strong real-world software engineering capability. This benchmark tests the model's ability to resolve actual GitHub issues.

How good is Qwen 3 Coder at math and reasoning?

Qwen 3 Coder scores 76.0% on the MATH benchmark (competition-level mathematics). It also achieves 48.0% on GPQA Diamond, a graduate-level science reasoning benchmark.

What is the context window of Qwen 3 Coder?

Qwen 3 Coder has a context window of 262K tokens. This determines how much text, conversation history, and code the model can process in a single request.

Who created Qwen 3 Coder?

Qwen 3 Coder was created by Qwen. It is classified as a open source model in our catalogue.

Is Qwen 3 Coder open source?

Yes, Qwen 3 Coder is an open-source model. The model weights are publicly available for download and self-hosting.