Qwen3 235B A22B Instruct 2507 vs Qwen3.5 397B A17B

Wins 0 of 4 benchmarks
$0.47 / task
Wins 1 of 4 benchmarks
Qwen3 235B A22B Instruct 250711.1
Qwen3.5 397B A17B29
Qwen3 235B A22B Instruct 2507—
Qwen3.5 397B A17B0
Qwen3 235B A22B Instruct 2507—
Qwen3.5 397B A17B6.9
Qwen3 235B A22B Instruct 2507—
Qwen3.5 397B A17B77.3

Specs

SpecQwen3 235B A22B Instruct 2507Qwen3.5 397B A17B
Input, per 1M$0.09$0.60
Output, per 1M$0.55$3.60
Cost per task?—$0.47
Speed, tok/s4958
Context262K262K
WeightsOpenOpen
ReleasedJul 2025Feb 2026
Memory (4-bit)141 GB242 GB
LicenceApache-2.0Apache-2.0
Commercial useAllowedAllowed
See images—✓
Read PDFs——
Hear audio——
Watch video—✓
Call tools✓✓
Force a tool callPartialPartial
Return strict JSON✓✓
Search the web??
Run code??
Use a computer??

What breaks when you switch

Qwen3 235B A22B Instruct 2507 → Qwen3.5 397B A17B
Memory (4-bit): 141 GB → 242 GB
Qwen3.5 397B A17B → Qwen3 235B A22B Instruct 2507
No image input