Model Comparator

Compare up to four models using the same criteria.

Qwen3.5 397B A17B (Non-reasoning)
Alibaba
Qwen3.6 27B (Reasoning)
Alibaba

Differences only

✓ Best published value among measured models. — Data unavailable. These metrics do not establish an overall ranking.

Model Comparator
Metrics	Qwen3.5 397B A17B (Non-reasoning) Alibaba	Qwen3.6 27B (Reasoning) Alibaba
Intelligence	21.4	21.4
Coding	—	53.7
Agentic	—	18.5
Speed	Best: 56.4 t/s	55.6 t/s
TTFT	Best: 2.43 s	3.62 s
Input · $ / 1M tokens	$0.60	$0.60
Output · $ / 1M tokens	$3.60	$3.60
Context window	262K	262K
Reasoning	No	Yes
Open weights	Yes	Yes
Cost / task	—	$0.621

Intelligence

Qwen3.5 397B A17B (Non-reasoning): 21.4
Qwen3.6 27B (Reasoning): 21.4

Coding

Qwen3.5 397B A17B (Non-reasoning): —
Qwen3.6 27B (Reasoning): 53.7

Agentic

Qwen3.5 397B A17B (Non-reasoning): —
Qwen3.6 27B (Reasoning): 18.5

Speed

Qwen3.5 397B A17B (Non-reasoning): Best: 56.4 t/s
Qwen3.6 27B (Reasoning): 55.6 t/s

TTFT

Qwen3.5 397B A17B (Non-reasoning): Best: 2.43 s
Qwen3.6 27B (Reasoning): 3.62 s

Input · $ / 1M tokens

Qwen3.5 397B A17B (Non-reasoning): $0.60
Qwen3.6 27B (Reasoning): $0.60

Output · $ / 1M tokens

Qwen3.5 397B A17B (Non-reasoning): $3.60
Qwen3.6 27B (Reasoning): $3.60

Context window

Qwen3.5 397B A17B (Non-reasoning): 262K
Qwen3.6 27B (Reasoning): 262K

Reasoning

Qwen3.5 397B A17B (Non-reasoning): No
Qwen3.6 27B (Reasoning): Yes

Open weights

Qwen3.5 397B A17B (Non-reasoning): Yes
Qwen3.6 27B (Reasoning): Yes

Cost / task

Qwen3.5 397B A17B (Non-reasoning): —
Qwen3.6 27B (Reasoning): $0.621

Each reasoning level remains a separate configuration.

Score and cost

Each line joins the reasoning levels of a compared model; the large point marks the configuration in the table. The top-right corner holds the best score for the lowest cost. Select another point to compare that level.

Qwen3.5 397B A17B (Non-reasoning)21.4 · $1.35
Qwen3.6 27B (Reasoning)21.4 · $1.35