Model Comparator

Compare up to four models using the same criteria.

Agnes 2.5 Pro Alpha (based on Qwen3.5 397B A17B)
Sapiens AI
Quasar 438B (Max, Based on GLM-5.2)
Multiverse Computing

Differences only

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

Model Comparator
Metrics	Agnes 2.5 Pro Alpha (based on Qwen3.5 397B A17B) Sapiens AI	Quasar 438B (Max, Based on GLM-5.2) Multiverse Computing
Intelligence	Best: 26.8	26.7
Coding	58.8	Best: 61.2
Agentic	—	31.4
Speed	160 t/s	Best: 185 t/s
TTFT	4.26 s	Best: 1.33 s
Input · $ / 1M tokens	$0.45×4.1 relative to the lowest price among the compared models	Best: $0.11
Output · $ / 1M tokens	$0.90×2.7 relative to the lowest price among the compared models	Best: $0.33
Context window	1M	1M
Reasoning	Yes	Yes
Open weights	Yes	No
Cost / task	—	$2.02

Intelligence

Agnes 2.5 Pro Alpha (based on Qwen3.5 397B A17B): Best: 26.8
Quasar 438B (Max, Based on GLM-5.2): 26.7

Coding

Agnes 2.5 Pro Alpha (based on Qwen3.5 397B A17B): 58.8
Quasar 438B (Max, Based on GLM-5.2): Best: 61.2

Agentic

Agnes 2.5 Pro Alpha (based on Qwen3.5 397B A17B): —
Quasar 438B (Max, Based on GLM-5.2): 31.4

Speed

Agnes 2.5 Pro Alpha (based on Qwen3.5 397B A17B): 160 t/s
Quasar 438B (Max, Based on GLM-5.2): Best: 185 t/s

TTFT

Agnes 2.5 Pro Alpha (based on Qwen3.5 397B A17B): 4.26 s
Quasar 438B (Max, Based on GLM-5.2): Best: 1.33 s

Input · $ / 1M tokens

Agnes 2.5 Pro Alpha (based on Qwen3.5 397B A17B): $0.45×4.1 relative to the lowest price among the compared models
Quasar 438B (Max, Based on GLM-5.2): Best: $0.11

Output · $ / 1M tokens

Agnes 2.5 Pro Alpha (based on Qwen3.5 397B A17B): $0.90×2.7 relative to the lowest price among the compared models
Quasar 438B (Max, Based on GLM-5.2): Best: $0.33

Context window

Agnes 2.5 Pro Alpha (based on Qwen3.5 397B A17B): 1M
Quasar 438B (Max, Based on GLM-5.2): 1M

Reasoning

Agnes 2.5 Pro Alpha (based on Qwen3.5 397B A17B): Yes
Quasar 438B (Max, Based on GLM-5.2): Yes

Open weights

Agnes 2.5 Pro Alpha (based on Qwen3.5 397B A17B): Yes
Quasar 438B (Max, Based on GLM-5.2): No

Cost / task

Agnes 2.5 Pro Alpha (based on Qwen3.5 397B A17B): —
Quasar 438B (Max, Based on GLM-5.2): $2.02

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.

Agnes 2.5 Pro Alpha (based on Qwen3.5 397B A17B)26.8 · $0.563
Quasar 438B (Max, Based on GLM-5.2)26.7 · $0.165