Claude Sonnet 4.5 vs Qwen3-Max

Benchmarks, API pricing and specs, head to head. Data updated 2026-06-10.

Claude Sonnet 4.5

Anthropic · Sep 2025

65.5

The model that made 30-hour autonomous coding sessions real — 77.2% SWE-bench Verified and a 1M-token context beta at $3/$15.

Qwen3-Max

Alibaba (Qwen) · Sep 2025

80.5

Alibaba's trillion-parameter API flagship — frontier-adjacent quality with strong agentic tool use at mid-tier prices.

The verdict

Claude Sonnet 4.5 wins 2 of the 3 benchmarks these models share, against 1 for Qwen3-Max. Qwen3-Max is about 2.5x cheaper per blended million tokens (3:1 input:output mix). Qwen3-Max also takes 262K of context versus 200K for Claude Sonnet 4.5.

Benchmark head-to-head 21

SWE-bench Verified
77.2%69.6%
Claude Sonnet 4.5Qwen3-Max
GPQA Diamond
83.4%~85.4%
Claude Sonnet 4.5Qwen3-Max
AIME 2025
87%~81.6%
Claude Sonnet 4.5Qwen3-Max

Specs & pricing

Claude Sonnet 4.5Qwen3-Max
modhub Index65.580.5
Input price / 1M$3$1.2
Output price / 1M$15$6
Context window200K262K
Max output64K66K
Open weightsnono
Reasoning modelyesyes
Multimodal inputtext, imagetext
Knowledge cutoffJan 2025Jun 2025
ReleasedSep 2025Sep 2025
Example monthly cost*$52.50$21.00

* 10M input + 1.5M output tokens per month at list prices, no caching. Green = better value on that row.

Frequently asked questions

Which is better, Claude Sonnet 4.5 or Qwen3-Max?
Claude Sonnet 4.5 wins 2 of the 3 benchmarks these models share, against 1 for Qwen3-Max. Qwen3-Max is about 2.5x cheaper per blended million tokens (3:1 input:output mix). Qwen3-Max also takes 262K of context versus 200K for Claude Sonnet 4.5.
Which is cheaper, Claude Sonnet 4.5 or Qwen3-Max?
Claude Sonnet 4.5 costs $3/$15 per million input/output tokens, while Qwen3-Max costs $1.2/$6. For a typical workload of 10M input and 1.5M output tokens per month, that's $52.50 versus $21.00.
Which model is better for coding, Claude Sonnet 4.5 or Qwen3-Max?
On SWE-bench Verified — the standard agentic-coding benchmark — Claude Sonnet 4.5 scores 77.2% versus 69.6% for Qwen3-Max, making Claude Sonnet 4.5 the stronger pick for coding agents.

More comparisons