prices synced 2026-07-28

Qwen3 235B A22B Instruct 2507

Alibaba · Released Jul 2025

Compare

A multilingual mixture-of-experts model with 235B total parameters and 22B active per forward pass, instruction-tuned for general-purpose text generation.

Strengths
Efficient inference through sparse activation with a large effective capacity; supports multilingual tasks across diverse languages.
Best for
Text generation, instruction following, and multilingual applications where latency and throughput matter alongside capability.
Limitations
Mixture-of-experts routing adds complexity; performance depends on workload alignment with the expert specialization; may not match dense models of comparable active parameter count on all tasks.

Input / 1M

$0.09

Output / 1M

$0.55

Cached input / 1M

Context window

262K

Price history

Input (solid)Output (dashed)
Price per 1M tokens over timeOutput went from $0.1 to $0.55; Input went from $0.09 to $0.09 between Jun 2026 and Jul 2026.$0$0.2$0.4$0.6Output on 11 Jun 2026: $0.1Output on 12 Jul 2026: $0.55Out $0.55Input on 11 Jun 2026: $0.09Input on 12 Jul 2026: $0.09In $0.09Jun 2026Jul 2026

Price change

30d
in decreased 0.0% out increased 450.0%
90d
in decreased 0.0% out increased 450.0%
1y
in decreased 0.0% out increased 450.0%
Since launch
in decreased 0.0% out increased 450.0%

Snapshots

Effective Input Output Cached in Note Source
12 Jul 2026 $0.09 $0.55 Imported from OpenRouter openrouter.ai
11 Jun 2026 $0.09 $0.1 Imported from OpenRouter openrouter.ai

More from Alibaba

Report a problem