MB / MODELS

Model leaderboard

Ranked by daily token usage across 539 models

Data date 2026-09-12 Snapshot data Last synced 14:58 Back to home

Full leaderboard 539 models · page 4 of 12

#	Model	Context	Daily tokens	Daily requests	Daily change
148	qwen/qwen3.5-122b-a10b-20260224	—	13.9B	1.8M	56.2%
149	liquid/lfm-2.5-2.6b-20260811 free	—	13.8B	1.3M	19.0%
150	meta/muse-spark-1.1-20260709	—	13.6B	803.3K	22.8%
151	qwen/qwen3-embedding-4b	—	13.6B	5.3M	19.5%
152	anthropic/claude-4.5-opus-20251124	—	13B	1M	29.9%
153	mistralai/mistral-small-24b-instruct-2501	—	12.6B	4M	1.3%
154	google/gemini-embedding-001	—	12.4B	16.9M	76.7%
155	google/gemma-3-12b-it	—	12.2B	8.7M	9.9%
156	openai/gpt-4o-mini-2024-07-18	—	12.1B	8.6M	57.9%
157	deepseek/deepseek-r1-0528	—	12B	1.3M	8.2%
158	openai/gpt-4o	—	11.9B	5.7M	16.2%
159	ibm-granite/granite-4.0-h-micro	—	11.5B	3.1M	428.4%
160	z-ai/glm-4.6	—	11.4B	843.9K	15.9%
161	openai/gpt-5-2025-08-07	—	11.2B	2M	3.1%
162	qwen/qwen3.5-27b-20260224	—	10.9B	3.1M	4.3%
163	voyageai/rerank-2.5-20260727	—	10B	2.3M	6.8%
164	qwen/qwen3-vl-8b-instruct	—	10B	10M	32.4%
165	nvidia/nemotron-3-super-120b-a12b-20230311	—	9.9B	1.1M	18.6%
166	aion-labs/aion-3.0-20260707	—	9.7B	479K	0.2%
167	bytedance-seed/seed-2.0-mini-20260224	—	9.1B	2.1M	26.6%
168	qwen/qwen3-vl-30b-a3b-instruct	—	8.9B	8M	3.4%
169	qwen/qwen3-8b-04-28	—	8.9B	2.8M	20.6%
170	qwen/qwen3-coder-30b-a3b-instruct	—	8.7B	839.4K	57.1%
171	qwen/qwen3-next-80b-a3b-instruct-2509	—	8.7B	1.8M	33.5%
172	meituan/longcat-2.0-20260720	—	8.5B	313.9K	44.8%
173	qwen/qwen3-reranker-8b	—	8.3B	648.2K	97.4%
174	mistralai/mistral-large-2512	—	7.9B	2.8M	49.5%
175	openai/gpt-oss-safeguard-20b	—	7.7B	3.3M	3.7%
176	deepseek/deepseek-v3.1-terminus	—	7.4B	840.7K	0.8%
177	z-ai/glm-4.5-air	—	6.5B	1.2M	63.3%
178	tencent/hy3-preview-20260421	—	6.3B	942.9K	8.7%
179	nvidia/nemotron-3-nano-30b-a3b	—	6.2B	884.6K	39.3%
180	thinkingmachines/inkling-small-20260730	—	6.2B	990.5K	44.4%
181	poolside/laguna-xs-2.1-20260625	—	6.2B	243.5K	76.5%
182	openai/gpt-5.6-luna-20260709 batch	—	5.8B	1.5M	22.5%
183	mistralai/mistral-embed-2312	—	5.6B	906.8K	20.1%
184	sao10k/l3-lunaris-8b	—	5.5B	2.3M	16.5%
185	qwen/qwen3-coder-flash	—	5.1B	172.9K	190.9%
186	z-ai/glm-5v-turbo-20260401	—	5B	259.4K	21.0%
187	qwen/qwen-2.5-7b-instruct	—	4.9B	5.1M	10.6%
188	mistralai/ministral-8b-2512	—	4.8B	3.6M	12.7%
189	amazon/nova-lite-v1	—	4.6B	1.7M	63.8%
190	mistralai/ministral-3b-2512	—	4.1B	3.2M	97.1%
191	google/gemini-3.7-flash-20260813 batch	—	4.1B	585.7K	64.4%
192	google/gemini-3.8-flash-20260902 batch	—	4B	794.5K	216.5%
193	qwen/qwen3-max	—	3.7B	795.4K	42.7%
194	moonshotai/kimi-k2-0905	—	3.6B	1.1M	0.2%
195	google/gemini-3.1-flash-image-20260528	—	3.5B	1.3M	2.6%

Daily usage = prompt + completion + reasoning tokens; change is versus the previous day. No public source link is exposed yet; treat the data as reference only.