AI Battle > Rankings > Model Size: Total and Active Parameters

Model Size: Total and Active Parameters Ranking — LLM Models (2026)

Kimi K3 (max) by Kimi leads the Model Size: Total and Active Parameters ranking with a score of 104. The ranking covers 14 evaluated llm models, re-scored on every AI Battle telemetry run.

Last updated: 2026-09-14. Source: AI Battle live evaluation telemetry, 14 models ranked.

#	Model	Creator	Score
1	Kimi K3 (max)	Kimi	104
2	Qwen3.8 2.4T A95B	Alibaba	95.0
3	DeepSeek V4 Pro 0813 (Reasoning, Max Effort)	DeepSeek	49.0
4	Inkling (xhigh)	Thinking Machines	41.0
5	GLM-5.3 (max)	Z AI	40.0
6	DeepSeek V4.1 Flash (Reasoning, Max Effort)	DeepSeek	16.0
7	Nemotron 3 Ultra 550B A55B (Reasoning)	NVIDIA	55.0
8	MiniMax-M3	MiniMax	23.0
9	K2 Horizon 375B A23B	MBZUAI Institute of Foundation Models	23.0
10	GLM-5.3-Flash	Z AI	18.0

FAQ — Model Size: Total and Active Parameters

What is the best model for Model Size: Total and Active Parameters in 2026?

Kimi K3 (max) by Kimi currently leads the Model Size: Total and Active Parameters ranking with a score of 104, based on live AI Battle telemetry across LLM Models.

How often is the Model Size: Total and Active Parameters ranking updated?

The Model Size: Total and Active Parameters ranking is refreshed automatically from the AI Battle evaluation cluster. Scores are re-evaluated on every telemetry run, so positions reflect the latest published results.

Which models rank in the top 3 for Model Size: Total and Active Parameters?

The top 3 are: 1. Kimi K3 (max) (104), 2. Qwen3.8 2.4T A95B (95.0), 3. DeepSeek V4 Pro 0813 (Reasoning, Max Effort) (49.0).

All AI Battle Rankings

• Artificial Analysis Intelligence Index — LLM Models ranking
• End-to-End Response Time — LLM Models ranking
• AA-Omniscience Index — LLM Models ranking
• Artificial Analysis Intelligence Index by Open Weights / Proprietary — LLM Models ranking
• AA-Briefcase Elo — LLM Models ranking
• Cost per Intelligence Index Task — LLM Models ranking
• Cost to Run Artificial Analysis Intelligence Index — LLM Models ranking
• Output Tokens per Intelligence Index Task — LLM Models ranking
• Pricing: Cache Hit, Input, and Output — LLM Models ranking
• Output Speed — LLM Models ranking
• Time per Intelligence Index Task — LLM Models ranking
• Context Window — LLM Models ranking
• Latency: Time To First Answer Token — LLM Models ranking
• Artificial Analysis Openness Index: Score — LLM Models ranking
• Speed — LLM Models ranking
• Cost per Task — LLM Models ranking
• Intelligence — LLM Models ranking
• GDPval-AA v2 Leaderboard — LLM Index ranking
• Artificial Analysis Intelligence Index by Open Weights / Proprietary — LLM Index ranking
• Cost per Intelligence Index Task — LLM Index ranking
• Artificial Analysis Intelligence Index — LLM Index ranking
• Artificial Analysis Finance & Accounting Index — LLM Index ranking
• AA-Omniscience Index — LLM Index ranking
• AA-Briefcase Elo — LLM Index ranking