Model Comparator

Compare up to four models using the same criteria.

Claude Haiku 5.5 (Medium)
Anthropic
JT-4.1 Flash 236B A21B (Reasoning)
China Mobile

Differences only

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

Model Comparator
Metrics	Claude Haiku 5.5 (Medium) Anthropic	JT-4.1 Flash 236B A21B (Reasoning) China Mobile
Intelligence	Best: 34.5	33.9
Coding	—	52.4
Speed	144 t/s	—
TTFT	12 s	—
Input · $ / 1M tokens	$0.10	—
Output · $ / 1M tokens	$0.50	—
Context window	Best: 1M	262K
Reasoning	Yes	Yes
Open weights	No	No
Cost / task	$0.047	—

Intelligence

Claude Haiku 5.5 (Medium): Best: 34.5
JT-4.1 Flash 236B A21B (Reasoning): 33.9

Coding

Claude Haiku 5.5 (Medium): —
JT-4.1 Flash 236B A21B (Reasoning): 52.4

Speed

Claude Haiku 5.5 (Medium): 144 t/s
JT-4.1 Flash 236B A21B (Reasoning): —

TTFT

Claude Haiku 5.5 (Medium): 12 s
JT-4.1 Flash 236B A21B (Reasoning): —

Input · $ / 1M tokens

Claude Haiku 5.5 (Medium): $0.10
JT-4.1 Flash 236B A21B (Reasoning): —

Output · $ / 1M tokens

Claude Haiku 5.5 (Medium): $0.50
JT-4.1 Flash 236B A21B (Reasoning): —

Context window

Claude Haiku 5.5 (Medium): Best: 1M
JT-4.1 Flash 236B A21B (Reasoning): 262K

Reasoning

Claude Haiku 5.5 (Medium): Yes
JT-4.1 Flash 236B A21B (Reasoning): Yes

Open weights

Claude Haiku 5.5 (Medium): No
JT-4.1 Flash 236B A21B (Reasoning): No

Cost / task

Claude Haiku 5.5 (Medium): $0.047
JT-4.1 Flash 236B A21B (Reasoning): —

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.

Claude Haiku 5.5 (Medium)34.5 · $0.047

Not plotted for lack of data on these axes: JT-4.1 Flash 236B A21B (Reasoning).