LLMs

LLMs ranking

298 published · sponsored rows sit above the ranking

Rank	Model	Developer	Origin	$ / 1M in / out	Context	Open weights	Last changed
Ranking · 298 match
221 —	Falcon3-10B-InstructTII model · 4 versions	TII	AE	—	—	Yes
222 —	GELab-Zero-4B-previewStepFun model	StepFun	CN	—	—	Yes
223 —	granite-3.0-1b-a400m-instructIBM model	IBM	US	—	—	Yes
224 —	MiMo-VL-7B-RL-2508Xiaomi model	Xiaomi	CN	—	—	Yes
225 —	Phi-4-mini-instructMicrosoft model	Microsoft	US	—	—	Yes
226 New	DeepSeek: DeepSeek V4.1 Flash 1017DeepSeek model	DeepSeek	CN	Free	1M	Yes	22 OctDeepSeek V4.1 Flash 1017: Sparse MoE model with native image understanding. Opti
227 —	MiniMax-VL-01MiniMax model	MiniMax	CN	—	—	Yes
228 —	Kimi-VL-A3B-Thinking-2506Moonshot AI model · 2 versions	Moonshot AI	CN	—	—	Yes
229 —	Nemotron 3 Nano 30B A3BNVIDIA model	NVIDIA	US	$0.06 / $0.24	262K	Yes	20 OctNemotron 3 Nano 30B A3B is a powerful LLM for various NLP tasks.
230 —	GLM 4.5 AirZhipu AI model	Zhipu AI	CN	$0.13 / $0.85	131K	Yes
231 —	Llama-2-13b-chat-hfMeta model · 3 versions	Meta	US	—	—	Yes
232 New	Mathstral 7BMistral AI model	Mistral AI	FR	—	—	Yes
233 —	Qwen: Qwen3.6 PlusAlibaba model	Alibaba	CN	$0.32	1M	Yes	21 OctQwen3.6 Plus offers major gains in coding, reasoning, and multimodal capabilitie
234 —	solar-pro-preview-instructUpstage model	Upstage	KR	—	—	Yes
235 —	tiny-aya-globalCohere model	Cohere	CA	—	—	Yes
236 —	Step 5 PreviewStepFun model	StepFun	CN	$1 / $2.70	1M	Yes
237 —	DeepSeek: R1 0528DeepSeek model	DeepSeek	CN	$0.50	163K	Yes	22 OctDeepSeek R1 0528: Open-source, 671B params, 164K context, $0.50/$2.15 per 1M tok
238 —	Kimi-Linear-48B-A3B-InstructMoonshot AI model	Moonshot AI	CN	—	—	Yes
239 —	Nemotron 3 Nano OmniNVIDIA model	NVIDIA	US	$0 / $0	256K	Yes
240 —	GLM 4.7 FlashZhipu AI model	Zhipu AI	CN	$0.06 / $0.40	200K	Yes

Latest in LLMs verified changes

DeepSeek R1 0528: Open-source, 671B params, 164K context, $0.50/$2.15 per 1M tokens.

DeepSeek V4.1 Flash 1017: Sparse MoE model with native image understanding. Optimized for coding and terminal use.

DeepSeek V4 Pro 0813 model released with 1M context. Pricing is $0.2235/1M input and $8/1M output.

DeepSeek V4 Pro 0423: 1.6T params, 1M context, advanced reasoning, coding. $0.45/$3.48 per token.

Qwen-Plus: 131K context, balanced performance, speed, and cost. Released Feb 1, 2025.

LLMs ranking