Roundup
Best AI Models for Coding
Which frontier models actually hold up in multi-file refactors, agentic tooling and test-driven work, and what they cost per million tokens.
qwen/qwen3-235b-a22b-thinking-2507
Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks. It activates 22B of its 235B parameters per forward pass and natively supports up to 262,144...
Source: OpenRouter (live)
Prices as listed on OpenRouter, refreshed every five minutes. Provider-direct prices may differ. Convert to your currency on the AI Tracker.
Capabilities
Capabilities are read from the provider listing. “Not declared” means the listing does not advertise the feature, not that it is absent.
Why it costs this
Alternatives
Qwen (Alibaba)
$3.00
/M blended
Qwen (Alibaba)
$0.230
/M blended
Qwen (Alibaba)
$0.788
/M blended
Qwen (Alibaba)
$3.00
/M blended
Qwen: Qwen3.8 2.4T A95B (batch)
Qwen (Alibaba)
$3.00
/M blended
Qwen (Alibaba)
$0.055
/M blended
Alternatives
$0.750
/M blended
$0.750
/M blended
$0.750
/M blended
OpenAI
$0.750
/M blended
Z.ai (Zhipu)
$0.738
/M blended
Z.ai (Zhipu)
$0.760
/M blended
Within ±50% of $0.747/M blended.
Analysis
Roundup
Which frontier models actually hold up in multi-file refactors, agentic tooling and test-driven work, and what they cost per million tokens.
Alibaba Cloud's Qwen open-weight family. All Qwen (Alibaba) models · Official site
Newsletter
Get the most important developments in artificial intelligence delivered directly to your inbox.
No hype. No spam.
Just trusted insights, major model releases, pricing updates, benchmark changes, product reviews, and practical guidance from across the AI ecosystem.
The AI HQ Briefing
One email a week. Read in five minutes.