Model Comparator

Compare up to four models using the same criteria.

Qwen3 Max
Alibaba
Claude 4.5 Haiku (Non-reasoning)
Anthropic

Differences only

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

Model Comparator
Metrics	Qwen3 Max Alibaba	Claude 4.5 Haiku (Non-reasoning) Anthropic
Intelligence	Best: 15.6	15.4
Coding	—	43.9
Math	Best: 80.7	39
Agentic	—	8
Speed	61.2 t/s	Best: 87.3 t/s
TTFT	2.25 s	Best: 0.5 s
Input · $ / 1M tokens	$1.20×1.2 relative to the lowest price among the compared models	Best: $1
Output · $ / 1M tokens	$6×1.2 relative to the lowest price among the compared models	Best: $5
Context window	Best: 262K	200K
Reasoning	No	No
Open weights	No	No

Intelligence

Qwen3 Max: Best: 15.6
Claude 4.5 Haiku (Non-reasoning): 15.4

Coding

Qwen3 Max: —
Claude 4.5 Haiku (Non-reasoning): 43.9

Math

Qwen3 Max: Best: 80.7
Claude 4.5 Haiku (Non-reasoning): 39

Agentic

Qwen3 Max: —
Claude 4.5 Haiku (Non-reasoning): 8

Speed

Qwen3 Max: 61.2 t/s
Claude 4.5 Haiku (Non-reasoning): Best: 87.3 t/s

TTFT

Qwen3 Max: 2.25 s
Claude 4.5 Haiku (Non-reasoning): Best: 0.5 s

Input · $ / 1M tokens

Qwen3 Max: $1.20×1.2 relative to the lowest price among the compared models
Claude 4.5 Haiku (Non-reasoning): Best: $1

Output · $ / 1M tokens

Qwen3 Max: $6×1.2 relative to the lowest price among the compared models
Claude 4.5 Haiku (Non-reasoning): Best: $5

Context window

Qwen3 Max: Best: 262K
Claude 4.5 Haiku (Non-reasoning): 200K

Reasoning

Qwen3 Max: No
Claude 4.5 Haiku (Non-reasoning): No

Open weights

Qwen3 Max: No
Claude 4.5 Haiku (Non-reasoning): No

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 Max15.6 · $2.40
Claude 4.5 Haiku (Non-reasoning)15.4 · $2