Training & Fine-Tuning Cost Leaderboard

Find the cheapest way to train or fine-tune any AI model.

250 models · 14 methods · 60 GPUs

Rank▲	Model	Params	Method	GPU	GPUs	Time	Cost	Quality	Actions
🥇	Qwen 2.5 7B Qwen 2.5	7.6B	QLoRA	T4	1	6.7h	$0.80	88%	Fine-tune →
🥈	Qwen 3 8B Qwen 3	8.2B	QLoRA	T4	1	7.2h	$0.86	88%	Fine-tune →
🥉	Qwen 2.5 1.5B Qwen 2.5	1.5B	GaLore	T4	1	1.3h	$0.16	97%	Fine-tune →
4	Qwen 2.5 3B Qwen 2.5	3.1B	GaLore	T4	1	2.7h	$0.32	97%	Fine-tune →
5	Llama 3.1 8B Llama 3.1	8B	QLoRA	T4	1	7.0h	$0.84	88%	Fine-tune →
6	Qwen 3 4B Qwen 3	4B	ReFT	T4	1	3.5h	$0.42	83%	Fine-tune →
7	Llama 3.2 3B Llama 3.2	3.2B	ReFT	T4	1	2.8h	$0.34	83%	Fine-tune →
8	Qwen 3 32B Qwen 3	32.8B	ReFT	H100 SXM	1	1.9h	$3.6	83%	Fine-tune →
9	Llama 3.2 1B Llama 3.2	1.2B	GaLore	T4	1	1.1h	$0.13	97%	Fine-tune →
10	Llama 3 8B Llama 3	8B	QLoRA	T4	1	7.0h	$0.84	88%	Fine-tune →
11	Llama 3.1 70B Llama 3.1	70.6B	QLoRA	H100 SXM	1	4.1h	$7.7	88%	Fine-tune →
12	HelpSteer2 Llama 3.1 70B Llama 3.1	70.6B	QLoRA	H100 SXM	1	4.1h	$7.7	88%	Fine-tune →
13	Llama 3.1 70B Turbo Llama 3.1	70.6B	QLoRA	H100 SXM	1	4.1h	$7.7	88%	Fine-tune →
14	Gemma 3 1B Gemma 3	1B	NEFTune	T4	1	53m	$0.11	85%	Fine-tune →
15	E5 Mistral 7B E5	7.1B	QLoRA	T4	1	6.2h	$0.75	88%	Fine-tune →
16	BioMistral 7B BioMistral	7.2B	QLoRA	T4	1	6.3h	$0.76	88%	Fine-tune →
17	NV EmbedQA Mistral 7B NV EmbedQA	7.2B	QLoRA	T4	1	6.3h	$0.76	88%	Fine-tune →
18	Mistral 7B Mistral	7.3B	QLoRA	T4	1	6.4h	$0.77	88%	Fine-tune →
19	TinyLlama 1.1B TinyLlama	1.1B	NEFTune	T4	1	58m	$0.12	85%	Fine-tune →
20	TinyLlama 1.1B Chat TinyLlama	1.1B	NEFTune	T4	1	58m	$0.12	85%	Fine-tune →
21	Qwen 2.5 14B Qwen 2.5	14.8B	QLoRA	T4	1	13h	$1.6	88%	Fine-tune →
22	Qwen 2.5 72B Qwen 2.5	72.7B	QLoRA	H100 SXM	1	4.2h	$7.9	88%	Fine-tune →
23	Phi 2 Phi	2.7B	GaLore	T4	1	2.4h	$0.28	97%	Fine-tune →
24	Qwen 2.5 32B Qwen 2.5	32.5B	ReFT	H100 SXM	1	1.9h	$3.5	83%	Fine-tune →
25	DeepSeek R1 Distill 1.5B DeepSeek R1	1.5B	GaLore	T4	1	1.3h	$0.16	97%	Fine-tune →
26	DeepSeek R1 Distill 8B DeepSeek R1	8B	QLoRA	T4	1	7.0h	$0.84	88%	Fine-tune →
27	DeepSeek R1 Distill 14B DeepSeek R1	14.8B	QLoRA	T4	1	13h	$1.6	88%	Fine-tune →
28	DeepSeek R1 Distill 32B DeepSeek R1	32.8B	ReFT	H100 SXM	1	1.9h	$3.6	83%	Fine-tune →
29	DeepSeek R1 Distill 70B DeepSeek R1	70.6B	QLoRA	H100 SXM	1	4.1h	$7.7	88%	Fine-tune →
30	DeepSeek R1 DeepSeek R1	671B MoE (37B active)	QLoRA	T4	31	24h	$91	88%	Fine-tune →
31	DeepSeek V3 DeepSeek V3	671B MoE (37B active)	QLoRA	T4	31	24h	$91	88%	Fine-tune →
32	Llama 3 70B 1M Context Llama 3	70.6B	QLoRA	H100 SXM	1	4.1h	$7.7	88%	Fine-tune →
33	Llama 3 70B Llama 3	70.6B	QLoRA	H100 SXM	1	4.1h	$7.7	88%	Fine-tune →
34	Mixtral 8x7B Mixtral	46.7B MoE (12.9B active)	QLoRA	H100 SXM	1	2.7h	$5.1	88%	Fine-tune →
35	Mixtral 8x7B Instruct Mixtral	46.7B MoE (12.9B active)	QLoRA	H100 SXM	1	2.7h	$5.1	88%	Fine-tune →
36	Gemma 2 9B Gemma 2	9.2B	QLoRA	T4	1	8.1h	$0.97	88%	Fine-tune →
37	Phi 4 Mini Phi	3.8B	ReFT	T4	1	3.3h	$0.40	83%	Fine-tune →
38	Phi-4 Phi	14.7B	QLoRA	T4	1	13h	$1.5	88%	Fine-tune →
39	Llama 2 7B Llama 2	7B	QLoRA	T4	1	6.1h	$0.74	88%	Fine-tune →
40	Llama 2 13B Llama 2	13B	QLoRA	T4	1	11h	$1.4	88%	Fine-tune →
41	Llama 2 70B Llama 2	70B	QLoRA	H100 SXM	1	4.0h	$7.6	88%	Fine-tune →
42	Llama Guard 3 1B Llama Guard	1B	NEFTune	T4	1	53m	$0.11	85%	Fine-tune →
43	Code Llama 7B Code Llama	7B	QLoRA	T4	1	6.1h	$0.74	88%	Fine-tune →
44	Llama Guard 3 8B Llama Guard	8B	QLoRA	T4	1	7.0h	$0.84	88%	Fine-tune →
45	Llama 3.3 8B Llama 3.3	8B	QLoRA	T4	1	7.0h	$0.84	88%	Fine-tune →
46	Llama 3.2 11B Vision Llama 3.2	11B	QLoRA	T4	1	9.6h	$1.2	88%	Fine-tune →
47	Code Llama 13B Code Llama	13B	QLoRA	T4	1	11h	$1.4	88%	Fine-tune →
48	Code Llama 34B Code Llama	34B	ReFT	H100 SXM	1	2.0h	$3.7	83%	Fine-tune →
49	Llama 3.1 Nemotron 51B Llama 3.1	51B	QLoRA	H100 SXM	1	2.9h	$5.5	88%	Fine-tune →
50	Code Llama 70B Code Llama	70B	QLoRA	H100 SXM	1	4.0h	$7.6	88%	Fine-tune →
51	Llama 3.3 70B Llama 3.3	70.6B	QLoRA	H100 SXM	1	4.1h	$7.7	88%	Fine-tune →
52	Llama 3.1 Nemotron 70B Instruct Llama 3.1	70.6B	QLoRA	H100 SXM	1	4.1h	$7.7	88%	Fine-tune →
53	Llama 3.1 Nemotron 70B Reward Llama 3.1	70.6B	QLoRA	H100 SXM	1	4.1h	$7.7	88%	Fine-tune →
54	Llama 3.2 90B Vision Instruct Llama 3.2	88.8B	QLoRA	H100 SXM	1	5.1h	$9.6	88%	Fine-tune →
55	Llama 3.2 90B Vision Llama 3.2	90B	QLoRA	H100 SXM	1	5.2h	$9.8	88%	Fine-tune →
56	Llama 4 Scout Llama 4	109B MoE (17B active)	QLoRA	B200 SXM	1	2.8h	$12	88%	Fine-tune →
57	Llama 4 Maverick Llama 4	400B MoE (17B active)	QLoRA	B200 NVL (pair)	1	5.1h	$53	88%	Fine-tune →
58	Llama 3.1 405B Llama 3.1	405B	QLoRA	B200 NVL (pair)	1	5.1h	$54	88%	Fine-tune →
59	Llama 4 Behemoth Llama 4	2000B MoE (400B active)	NEFTune	T4	1645	1.4h	$270	85%	Fine-tune →
60	Qwen 2.5 0.5B Qwen 2.5	500M	NEFTune	T4	1	26m	$0.05	85%	Fine-tune →
61	Qwen 3 0.6B Qwen 3	600M	NEFTune	T4	1	32m	$0.06	85%	Fine-tune →
62	Qwen 2.5 Coder 1.5B Qwen 2.5 Coder	1.5B	GaLore	T4	1	1.3h	$0.16	97%	Fine-tune →
63	Qwen 3 1.7B Qwen 3	1.7B	GaLore	T4	1	1.5h	$0.18	97%	Fine-tune →
64	Qwen 2 VL 2B Qwen 2 VL	2.2B	GaLore	T4	1	1.9h	$0.23	97%	Fine-tune →
65	Qwen 2.5 Coder 3B Qwen 2.5 Coder	3.1B	GaLore	T4	1	2.7h	$0.33	97%	Fine-tune →
66	GTE Qwen2 7B GTE	7.6B	QLoRA	T4	1	6.7h	$0.80	88%	Fine-tune →
67	Qwen 2 Audio 7B Qwen 2	7.6B	QLoRA	T4	1	6.7h	$0.80	88%	Fine-tune →
68	Qwen 2.5 Coder 7B Qwen 2.5 Coder	7.6B	QLoRA	T4	1	6.7h	$0.80	88%	Fine-tune →
69	Qwen 2.5 Math 7B Qwen 2.5 Math	7.6B	QLoRA	T4	1	6.7h	$0.80	88%	Fine-tune →
70	Qwen 2.5 VL 7B Qwen 2.5 VL	7.6B	QLoRA	T4	1	6.7h	$0.80	88%	Fine-tune →
71	Qwen 1.5 MoE A2.7B Qwen 1.5	14.3B MoE (2.7B active)	QLoRA	T4	1	13h	$1.5	88%	Fine-tune →
72	Qwen 2.5 Coder 14B Qwen 2.5 Coder	14.7B	QLoRA	T4	1	13h	$1.5	88%	Fine-tune →
73	Qwen 3 30B-A3B Qwen 3	30.5B MoE (3.3B active)	ReFT	H100 SXM	1	1.8h	$3.3	83%	Fine-tune →
74	Qwen 2.5 Coder 32B Qwen 2.5	32.5B	ReFT	H100 SXM	1	1.9h	$3.5	83%	Fine-tune →
75	Qwen 2.5 Math 72B Qwen 2.5 Math	72.7B	QLoRA	H100 SXM	1	4.2h	$7.9	88%	Fine-tune →
76	Qwen 2.5 VL 72B Qwen 2.5 VL	72.7B	QLoRA	H100 SXM	1	4.2h	$7.9	88%	Fine-tune →
77	Qwen 3 235B Qwen 3	235B MoE (22B active)	QLoRA	B200 SXM	1	5.9h	$27	88%	Fine-tune →
78	Codestral Mamba 7B Codestral	7.3B	QLoRA	T4	1	6.4h	$0.77	88%	Fine-tune →
79	Ministral 8B Ministral	8B	QLoRA	T4	1	7.0h	$0.84	88%	Fine-tune →
80	Mistral Nemo 12B Mistral Nemo	12B	QLoRA	T4	1	11h	$1.3	88%	Fine-tune →
81	Pixtral 12B Pixtral	12B	QLoRA	T4	1	11h	$1.3	88%	Fine-tune →
82	Codestral 22B Codestral	22B	ReFT	H100 SXM	1	1.3h	$2.4	83%	Fine-tune →
83	Mistral Small 24B Mistral Small	24B	ReFT	H100 SXM	1	1.4h	$2.6	83%	Fine-tune →
84	Mistral Small 3.1 24B Mistral Small	24B	ReFT	H100 SXM	1	1.4h	$2.6	83%	Fine-tune →
85	Mistral Medium 3 Mistral	70B	QLoRA	H100 SXM	1	4.0h	$7.6	88%	Fine-tune →
86	Mistral Large 2411 Mistral Large	123B	QLoRA	B200 SXM	1	3.1h	$14	88%	Fine-tune →
87	Mistral Large 2 Mistral Large	123B	QLoRA	B200 SXM	1	3.1h	$14	88%	Fine-tune →
88	Mixtral 8x22B Mixtral	141B MoE (39B active)	QLoRA	B200 SXM	1	3.6h	$16	88%	Fine-tune →
89	DeepSeek Coder 6.7B DeepSeek Coder	6.7B	ReFT	T4	1	5.9h	$0.70	83%	Fine-tune →
90	DeepSeek Math 7B DeepSeek Math	7.2B	QLoRA	T4	1	6.3h	$0.76	88%	Fine-tune →
91	DeepSeek V2 Lite DeepSeek V2	15.7B MoE (2.4B active)	QLoRA	T4	1	14h	$1.6	88%	Fine-tune →
92	DeepSeek MoE 16B DeepSeek MoE	16.4B MoE (2.8B active)	QLoRA	T4	1	14h	$1.7	88%	Fine-tune →
93	DeepSeek Coder 33B DeepSeek Coder	33B	ReFT	H100 SXM	1	1.9h	$3.6	83%	Fine-tune →
94	DeepSeek LLM 67B DeepSeek LLM	67B	QLoRA	H100 SXM	1	3.9h	$7.3	88%	Fine-tune →
95	DeepSeek Coder V2 236B DeepSeek Coder V2	236B MoE (21B active)	QLoRA	B200 SXM	1	6.0h	$27	88%	Fine-tune →
96	DeepSeek V2.5 DeepSeek V2	236B MoE (21B active)	QLoRA	B200 SXM	1	6.0h	$27	88%	Fine-tune →
97	Phi 1.5 Phi	1.3B	GaLore	T4	1	1.1h	$0.14	97%	Fine-tune →
98	Phi 1 Phi	1.3B	GaLore	T4	1	1.1h	$0.14	97%	Fine-tune →
99	Gemma 3 2B Gemma 3	2B	GaLore	T4	1	1.8h	$0.21	97%	Fine-tune →
100	Gemma 1.1 2B Gemma	2.5B	GaLore	T4	1	2.2h	$0.26	97%	Fine-tune →
101	Gemma 2 2B Gemma 2	2.6B	GaLore	T4	1	2.3h	$0.27	97%	Fine-tune →
102	RecurrentGemma 2B RecurrentGemma	2.7B	GaLore	T4	1	2.4h	$0.28	97%	Fine-tune →
103	PaLI-Gemma 3B PaLI-Gemma	2.9B	GaLore	T4	1	2.5h	$0.30	97%	Fine-tune →
104	Phi 3 Mini 3.8B Phi 3	3.8B	ReFT	T4	1	3.3h	$0.40	83%	Fine-tune →
105	Phi 3.5 Vision Phi 3.5	4.2B	ReFT	T4	1	3.7h	$0.44	83%	Fine-tune →
106	Gemma 3 4B Gemma 3	4.3B	ReFT	T4	1	3.8h	$0.45	83%	Fine-tune →
107	Phi 3 Small 7B Phi 3	7B	QLoRA	T4	1	6.1h	$0.74	88%	Fine-tune →
108	CodeGemma 7B Gemma	8.5B	QLoRA	T4	1	7.4h	$0.89	88%	Fine-tune →
109	Gemma 3 12B Gemma 3	12B	QLoRA	T4	1	11h	$1.3	88%	Fine-tune →
110	Phi 3 Medium 14B Phi 3	14B	QLoRA	T4	1	12h	$1.5	88%	Fine-tune →
111	Gemma 2 27B Gemma 2	27B	ReFT	H100 SXM	1	1.6h	$2.9	83%	Fine-tune →
112	Gemma 3 27B Gemma 3	27B	ReFT	H100 SXM	1	1.6h	$2.9	83%	Fine-tune →
113	Phi 3.5 MoE Phi	41.9B MoE (6.6B active)	QLoRA	H100 SXM	1	2.4h	$4.6	88%	Fine-tune →
114	Dolphin 2.9 72B Dolphin	72B	QLoRA	H100 SXM	1	4.1h	$7.8	88%	Fine-tune →
115	BGE Base EN v1.5 BGE	110M	NEFTune	T4	1	6m	$0.01	85%	Fine-tune →
116	SmolLM 135M SmolLM	135M	NEFTune	T4	1	7m	$0.01	85%	Fine-tune →
117	Nomic Embed Text v1.5 Nomic Embed	137M	NEFTune	T4	1	7m	$0.01	85%	Fine-tune →
118	Whisper Small Whisper	244M	NEFTune	T4	1	13m	$0.03	85%	Fine-tune →
119	NV EmbedQA E5 v5 NV EmbedQA	330M	NEFTune	T4	1	17m	$0.03	85%	Fine-tune →
120	NV Retriever v1 NV Retriever	330M	NEFTune	T4	1	17m	$0.03	85%	Fine-tune →
121	BGE Large EN v1.5 BGE	335M	NEFTune	T4	1	18m	$0.04	85%	Fine-tune →
122	SmolLM 360M SmolLM	360M	NEFTune	T4	1	19m	$0.04	85%	Fine-tune →
123	H2O Danube3 500M H2O Danube	500M	NEFTune	T4	1	26m	$0.05	85%	Fine-tune →
124	Cohere Embed English v3 Embed	500M	NEFTune	T4	1	26m	$0.05	85%	Fine-tune →
125	InfoXLM Large InfoXLM	550M	NEFTune	T4	1	29m	$0.06	85%	Fine-tune →
126	Multilingual E5 Large E5	560M	NEFTune	T4	1	29m	$0.06	85%	Fine-tune →
127	BGE M3 BGE	568M	NEFTune	T4	1	30m	$0.06	85%	Fine-tune →
128	Jina Embeddings v3 Jina Embeddings	570M	NEFTune	T4	1	30m	$0.06	85%	Fine-tune →
129	Parakeet CTC 0.6B Parakeet	600M	NEFTune	T4	1	32m	$0.06	85%	Fine-tune →
130	Whisper Medium Whisper	769M	NEFTune	T4	1	40m	$0.08	85%	Fine-tune →
131	Florence 2 Large Florence	770M	NEFTune	T4	1	40m	$0.08	85%	Fine-tune →
132	Canary 1B Canary	1B	NEFTune	T4	1	53m	$0.11	85%	Fine-tune →
133	Parakeet TDT 1.1B Parakeet	1.1B	NEFTune	T4	1	58m	$0.12	85%	Fine-tune →
134	SantaCoder 1.1B SantaCoder	1.1B	NEFTune	T4	1	58m	$0.12	85%	Fine-tune →
135	Whisper Large V3 Whisper	1.6B	GaLore	T4	1	1.4h	$0.16	97%	Fine-tune →
136	SmolLM2 1.7B SmolLM2	1.7B	GaLore	T4	1	1.5h	$0.18	97%	Fine-tune →
137	SeamlessM4T v2 Large SeamlessM4T	2.3B	GaLore	T4	1	2.0h	$0.24	97%	Fine-tune →
138	OpenELM 3B OpenELM	3B	GaLore	T4	1	2.6h	$0.32	97%	Fine-tune →
139	VILA 1.5 3B VILA	3B	GaLore	T4	1	2.6h	$0.32	97%	Fine-tune →
140	StableLM Zephyr 3B StableLM	3B	GaLore	T4	1	2.6h	$0.32	97%	Fine-tune →
141	BTLM 3B BTLM	3B	GaLore	T4	1	2.6h	$0.32	97%	Fine-tune →
142	StarCoder2 3B StarCoder2	3B	GaLore	T4	1	2.7h	$0.32	97%	Fine-tune →
143	Replit Code v1.5 3B Replit Code	3.3B	ReFT	T4	1	2.9h	$0.35	83%	Fine-tune →
144	DALL-E 3 DALL-E	3.5B	ReFT	T4	1	3.1h	$0.37	83%	Fine-tune →
145	Stable Diffusion XL 1.0 Stable Diffusion	3.5B	ReFT	T4	1	3.1h	$0.37	83%	Fine-tune →
146	Minitron 4B Nemotron	4B	ReFT	T4	1	3.5h	$0.42	83%	Fine-tune →
147	Nemotron Mini 4B Nemotron	4B	ReFT	T4	1	3.5h	$0.42	83%	Fine-tune →
148	ChatGLM3 6B ChatGLM3	6B	ReFT	T4	1	5.3h	$0.63	83%	Fine-tune →
149	Yi 6B 200K Yi	6B	ReFT	T4	1	5.3h	$0.63	83%	Fine-tune →
150	SciGLM 6B SciGLM	6.2B	ReFT	T4	1	5.4h	$0.65	83%	Fine-tune →
151	MPT 7B MPT	6.7B	ReFT	T4	1	5.9h	$0.70	83%	Fine-tune →
152	StarCoder2 7B StarCoder2	6.7B	ReFT	T4	1	5.9h	$0.71	83%	Fine-tune →
153	OLMo 2 7B OLMo 2	7B	QLoRA	T4	1	6.1h	$0.74	88%	Fine-tune →
154	Command R 7B Command R	7B	QLoRA	T4	1	6.1h	$0.74	88%	Fine-tune →
155	Falcon 7B Falcon	7B	QLoRA	T4	1	6.1h	$0.74	88%	Fine-tune →
156	Zephyr 7B Zephyr	7B	QLoRA	T4	1	6.1h	$0.74	88%	Fine-tune →
157	Vicuna 7B Vicuna	7B	QLoRA	T4	1	6.1h	$0.74	88%	Fine-tune →
158	OpenHermes 2.5 7B OpenHermes	7B	QLoRA	T4	1	6.1h	$0.74	88%	Fine-tune →
159	Baichuan 2 7B Baichuan 2	7B	QLoRA	T4	1	6.1h	$0.74	88%	Fine-tune →
160	SaulLM 7B SaulLM	7.2B	QLoRA	T4	1	6.3h	$0.76	88%	Fine-tune →
161	FinGPT 7B FinGPT	7.2B	QLoRA	T4	1	6.3h	$0.76	88%	Fine-tune →
162	Prometheus 2 7B Prometheus	7.2B	QLoRA	T4	1	6.3h	$0.76	88%	Fine-tune →
163	Falcon Mamba 7B Falcon Mamba	7.3B	QLoRA	T4	1	6.4h	$0.76	88%	Fine-tune →
164	Marco O1 Marco	7.6B	QLoRA	T4	1	6.7h	$0.80	88%	Fine-tune →
165	InternLM 2.5 7B InternLM 2.5	7.7B	QLoRA	T4	1	6.8h	$0.81	88%	Fine-tune →
166	NV Embed v2 NV Embed	7.8B	QLoRA	T4	1	6.9h	$0.82	88%	Fine-tune →
167	Aya 23 8B Aya	8B	QLoRA	T4	1	7.0h	$0.84	88%	Fine-tune →
168	Minitron 8B Nemotron	8B	QLoRA	T4	1	7.0h	$0.84	88%	Fine-tune →
169	GPT-4o Mini GPT-4	8B	QLoRA	T4	1	7.0h	$0.84	88%	Fine-tune →
170	Hermes 3 8B Hermes 3	8B	QLoRA	T4	1	7.0h	$0.84	88%	Fine-tune →
171	Yi Coder 9B Yi Coder	8.8B	QLoRA	T4	1	7.7h	$0.92	88%	Fine-tune →
172	Yi 1.5 9B Yi 1.5	8.8B	QLoRA	T4	1	7.7h	$0.93	88%	Fine-tune →
173	GLM-4 9B GLM-4	9.4B	QLoRA	T4	1	8.2h	$0.99	88%	Fine-tune →
174	ChatGLM4 9B ChatGLM	9.4B	QLoRA	T4	1	8.2h	$0.99	88%	Fine-tune →
175	SOLAR 10.7B SOLAR	10.7B	QLoRA	T4	1	9.4h	$1.1	88%	Fine-tune →
176	Falcon 11B Falcon	11B	QLoRA	T4	1	9.6h	$1.2	88%	Fine-tune →
177	Amazon Nova Lite Nova	12B	QLoRA	T4	1	11h	$1.3	88%	Fine-tune →
178	FLUX.1 Dev FLUX	12B	QLoRA	T4	1	11h	$1.3	88%	Fine-tune →
179	StableLM 2 12B StableLM 2	12.1B	QLoRA	T4	1	11h	$1.3	88%	Fine-tune →
180	KULLM 12.8B KULLM	12.8B	QLoRA	T4	1	11h	$1.3	88%	Fine-tune →
181	OLMo 2 13B OLMo 2	13B	QLoRA	T4	1	11h	$1.4	88%	Fine-tune →
182	Baichuan 2 13B Baichuan 2	13B	QLoRA	T4	1	11h	$1.4	88%	Fine-tune →
183	Vicuna 13B Vicuna	13B	QLoRA	T4	1	11h	$1.4	88%	Fine-tune →
184	Orca 2 13B Orca	13B	QLoRA	T4	1	11h	$1.4	88%	Fine-tune →
185	VILA 1.5 13B VILA	13B	QLoRA	T4	1	11h	$1.4	88%	Fine-tune →
186	ELYZA 13B ELYZA	13B	QLoRA	T4	1	11h	$1.4	88%	Fine-tune →
187	Cerebras GPT 13B Cerebras GPT	13B	QLoRA	T4	1	11h	$1.4	88%	Fine-tune →
188	Nekomata 14B Nekomata	14B	QLoRA	T4	1	12h	$1.5	88%	Fine-tune →
189	RWKV-6 14B RWKV	14.1B	QLoRA	T4	1	12h	$1.5	88%	Fine-tune →
190	Nemotron 15B Nemotron	15B	QLoRA	T4	1	13h	$1.6	88%	Fine-tune →
191	OctoCoder 15B OctoCoder	15.5B	QLoRA	T4	1	14h	$1.6	88%	Fine-tune →
192	StarCoder2 15B StarCoder2	15.5B	QLoRA	T4	1	14h	$1.6	88%	Fine-tune →
193	CodeGen2 16B CodeGen2	16B	QLoRA	T4	1	14h	$1.7	88%	Fine-tune →
194	CogVLM2 19B CogVLM2	19B	QLoRA	T4	1	17h	$2.0	88%	Fine-tune →
195	InternLM 2.5 20B InternLM 2.5	19.9B	QLoRA	T4	1	17h	$2.1	88%	Fine-tune →
196	GigaChat 20B GigaChat	20B	QLoRA	T4	1	18h	$2.1	88%	Fine-tune →
197	Claude 3.5 Haiku Claude	20B	QLoRA	T4	1	18h	$2.1	88%	Fine-tune →
198	GPT-3.5 Turbo GPT-3.5	20B	QLoRA	T4	1	18h	$2.1	88%	Fine-tune →
199	InternLM 20B InternLM	20B	QLoRA	T4	1	18h	$2.1	88%	Fine-tune →
200	Solar Pro 22B Solar	22B	ReFT	H100 SXM	1	1.3h	$2.4	83%	Fine-tune →
201	InternVL2 26B InternVL2	26B	ReFT	H100 SXM	1	1.5h	$2.8	83%	Fine-tune →
202	JAIS 30B JAIS	30B	ReFT	H100 SXM	1	1.7h	$3.3	83%	Fine-tune →
203	MPT 30B MPT	30B	ReFT	H100 SXM	1	1.7h	$3.3	83%	Fine-tune →
204	Vicuna 33B Vicuna	33B	ReFT	H100 SXM	1	1.9h	$3.6	83%	Fine-tune →
205	WizardCoder 33B WizardCoder	33B	ReFT	H100 SXM	1	1.9h	$3.6	83%	Fine-tune →
206	Yi 1.5 34B Yi 1.5	34.4B	ReFT	H100 SXM	1	2.0h	$3.7	83%	Fine-tune →
207	Aya 23 35B Aya	35B	QLoRA	H100 SXM	1	2.0h	$3.8	88%	Fine-tune →
208	Command R Command R	35B	QLoRA	H100 SXM	1	2.0h	$3.8	88%	Fine-tune →
209	Command R (August 2024) Command R	35B	QLoRA	H100 SXM	1	2.0h	$3.8	88%	Fine-tune →
210	Falcon 40B Falcon	40B	QLoRA	H100 SXM	1	2.3h	$4.3	88%	Fine-tune →
211	VILA 1.5 40B VILA	40B	QLoRA	H100 SXM	1	2.3h	$4.3	88%	Fine-tune →
212	Amazon Nova Pro Nova	50B	QLoRA	H100 SXM	1	2.9h	$5.4	88%	Fine-tune →
213	Gemini 2.0 Flash Gemini	50B MoE (15B active)	QLoRA	H100 SXM	1	2.9h	$5.4	88%	Fine-tune →
214	Gemini 1.5 Flash Gemini	50B MoE (12B active)	QLoRA	H100 SXM	1	2.9h	$5.4	88%	Fine-tune →
215	Jamba 1.5 Mini Jamba	52B	QLoRA	H100 SXM	1	3.0h	$5.6	88%	Fine-tune →
216	Jamba Instruct Jamba	52B MoE (12B active)	QLoRA	H100 SXM	1	3.0h	$5.6	88%	Fine-tune →
217	WizardMath 70B WizardMath	70B	QLoRA	H100 SXM	1	4.0h	$7.6	88%	Fine-tune →
218	Claude Sonnet 4 Claude	70B	QLoRA	H100 SXM	1	4.0h	$7.6	88%	Fine-tune →
219	o1-mini o1	70B	QLoRA	H100 SXM	1	4.0h	$7.6	88%	Fine-tune →
220	o3-mini o3	70B	QLoRA	H100 SXM	1	4.0h	$7.6	88%	Fine-tune →
221	Claude 3 Sonnet Claude	70B	QLoRA	H100 SXM	1	4.0h	$7.6	88%	Fine-tune →
222	Reka Core Reka	70B	QLoRA	H100 SXM	1	4.0h	$7.6	88%	Fine-tune →
223	Meditron 70B Meditron	70B	QLoRA	H100 SXM	1	4.0h	$7.6	88%	Fine-tune →
224	Japanese StableLM 70B StableLM	70B	QLoRA	H100 SXM	1	4.0h	$7.6	88%	Fine-tune →
225	Hermes 3 70B Hermes 3	70.6B	QLoRA	H100 SXM	1	4.1h	$7.7	88%	Fine-tune →
226	Nemotron 70B Nemotron	70.6B	QLoRA	H100 SXM	1	4.1h	$7.7	88%	Fine-tune →
227	Inflection 3 Inflection	100B	QLoRA	H100 SXM	1	5.7h	$11	88%	Fine-tune →
228	YaLM 100B YaLM	100B	QLoRA	H100 SXM	1	5.7h	$11	88%	Fine-tune →
229	Yi-Large Yi	103B MoE (24B active)	QLoRA	H100 SXM	1	5.9h	$11	88%	Fine-tune →
230	Command R+ Command R	104B	QLoRA	H100 SXM	1	6.0h	$11	88%	Fine-tune →
231	Command A Command	111B	QLoRA	B200 SXM	1	2.8h	$13	88%	Fine-tune →
232	DBRX Base DBRX	132B MoE (36B active)	QLoRA	B200 SXM	1	3.3h	$15	88%	Fine-tune →
233	DBRX Instruct DBRX	132B MoE (36B active)	QLoRA	B200 SXM	1	3.3h	$15	88%	Fine-tune →
234	Gemini 1.5 Pro Gemini	175B MoE (40B active)	QLoRA	B200 SXM	1	4.4h	$20	88%	Fine-tune →
235	Claude 3 Opus Claude	175B	QLoRA	B200 SXM	1	4.4h	$20	88%	Fine-tune →
236	Falcon 180B Falcon	180B	QLoRA	B200 SXM	1	4.6h	$20	88%	Fine-tune →
237	Claude Opus 4 Claude	200B	QLoRA	B200 SXM	1	5.1h	$23	88%	Fine-tune →
238	GPT-4o GPT-4	200B MoE (50B active)	QLoRA	B200 SXM	1	5.1h	$23	88%	Fine-tune →
239	GPT-4 Turbo GPT-4	200B MoE (50B active)	QLoRA	B200 SXM	1	5.1h	$23	88%	Fine-tune →
240	o1 o1	200B MoE (50B active)	QLoRA	B200 SXM	1	5.1h	$23	88%	Fine-tune →
241	Grok-2 Grok	314B MoE (50B active)	QLoRA	B200 NVL (pair)	1	4.0h	$42	88%	Fine-tune →
242	Nemotron 340B Nemotron	340B	QLoRA	B200 NVL (pair)	1	4.3h	$45	88%	Fine-tune →
243	Snowflake Arctic 128x3B Arctic	395B MoE (17B active)	QLoRA	B200 NVL (pair)	1	5.0h	$52	88%	Fine-tune →
244	Jamba 1.5 Large Jamba	398B	QLoRA	B200 NVL (pair)	1	5.0h	$53	88%	Fine-tune →
245	Snowflake Arctic 480B Arctic	480B MoE (17B active)	QLoRA	B200 NVL (pair)	1	6.1h	$64	88%	Fine-tune →
246	Megatron-Turing NLG 530B Megatron-Turing	530B	DoRA	T4	81	7.4h	$72	94%	Fine-tune →
247	Gemini 2.0 Pro Gemini	600B MoE (150B active)	NEFTune	T4	494	1.4h	$81	85%	Fine-tune →
248	Grok 3 Grok	600B MoE (120B active)	NEFTune	T4	494	1.4h	$81	85%	Fine-tune →
249	Kimi K2.5 Kimi	1000B MoE (32B active)	ReFT	T4	145	7.8h	$135	83%	Fine-tune →
250	GPT-4.5 Preview GPT	1500B MoE (300B active)	NEFTune	T4	1234	1.4h	$203	85%	Fine-tune →

Showing 250 of 250 models

🥇#1QLoRA

Qwen 2.5 7B

Qwen 2.5 · 7.6B

GPU

Time

6.7h

Cost

$0.80

Quality

88%

Fine-tune this model

🥈#2QLoRA

Qwen 3 8B

Qwen 3 · 8.2B

GPU

Time

7.2h

Cost

$0.86

Quality

88%

Fine-tune this model

🥉#3GaLore

Qwen 2.5 1.5B

Qwen 2.5 · 1.5B

GPU

Time

1.3h

Cost

$0.16

Quality

97%

Fine-tune this model

#4GaLore

Qwen 2.5 3B

Qwen 2.5 · 3.1B

GPU

Time

2.7h

Cost

$0.32

Quality

97%

Fine-tune this model

#5QLoRA

Llama 3.1 8B

Llama 3.1 · 8B

GPU

Time

7.0h

Cost

$0.84

Quality

88%

Fine-tune this model

#6ReFT

Qwen 3 4B

Qwen 3 · 4B

GPU

Time

3.5h

Cost

$0.42

Quality

83%

Fine-tune this model

#7ReFT

Llama 3.2 3B

Llama 3.2 · 3.2B

GPU

Time

2.8h

Cost

$0.34

Quality

83%

Fine-tune this model

#8ReFT

Qwen 3 32B

Qwen 3 · 32.8B

GPU

H100 SXM

Time

1.9h

Cost

$3.6

Quality

83%

Fine-tune this model

#9GaLore

Llama 3.2 1B

Llama 3.2 · 1.2B

GPU

Time

1.1h

Cost

$0.13

Quality

97%

Fine-tune this model

#10QLoRA

Llama 3 8B

Llama 3 · 8B

GPU

Time

7.0h

Cost

$0.84

Quality

88%

Fine-tune this model

#11QLoRA

Llama 3.1 70B

Llama 3.1 · 70.6B

GPU

H100 SXM

Time

4.1h

Cost

$7.7

Quality

88%

Fine-tune this model

#12QLoRA

HelpSteer2 Llama 3.1 70B

Llama 3.1 · 70.6B

GPU

H100 SXM

Time

4.1h

Cost

$7.7

Quality

88%

Fine-tune this model

#13QLoRA

Llama 3.1 70B Turbo

Llama 3.1 · 70.6B

GPU

H100 SXM

Time

4.1h

Cost

$7.7

Quality

88%

Fine-tune this model

#14NEFTune

Gemma 3 1B

Gemma 3 · 1B

GPU

Time

53m

Cost

$0.11

Quality

85%

Fine-tune this model

#15QLoRA

E5 Mistral 7B

E5 · 7.1B

GPU

Time

6.2h

Cost

$0.75

Quality

88%

Fine-tune this model

#16QLoRA

BioMistral 7B

BioMistral · 7.2B

GPU

Time

6.3h

Cost

$0.76

Quality

88%

Fine-tune this model

#17QLoRA

NV EmbedQA Mistral 7B

NV EmbedQA · 7.2B

GPU

Time

6.3h

Cost

$0.76

Quality

88%

Fine-tune this model

#18QLoRA

Mistral 7B

Mistral · 7.3B

GPU

Time

6.4h

Cost

$0.77

Quality

88%

Fine-tune this model

#19NEFTune

TinyLlama 1.1B

TinyLlama · 1.1B

GPU

Time

58m

Cost

$0.12

Quality

85%

Fine-tune this model

#20NEFTune

TinyLlama 1.1B Chat

TinyLlama · 1.1B

GPU

Time

58m

Cost

$0.12

Quality

85%

Fine-tune this model

#21QLoRA

Qwen 2.5 14B

Qwen 2.5 · 14.8B

GPU

Time

13h

Cost

$1.6

Quality

88%

Fine-tune this model

#22QLoRA

Qwen 2.5 72B

Qwen 2.5 · 72.7B

GPU

H100 SXM

Time

4.2h

Cost

$7.9

Quality

88%

Fine-tune this model

#23GaLore

Phi 2

Phi · 2.7B

GPU

Time

2.4h

Cost

$0.28

Quality

97%

Fine-tune this model

#24ReFT

Qwen 2.5 32B

Qwen 2.5 · 32.5B

GPU

H100 SXM

Time

1.9h

Cost

$3.5

Quality

83%

Fine-tune this model

#25GaLore

DeepSeek R1 Distill 1.5B

DeepSeek R1 · 1.5B

GPU

Time

1.3h

Cost

$0.16

Quality

97%

Fine-tune this model

#26QLoRA

DeepSeek R1 Distill 8B

DeepSeek R1 · 8B

GPU

Time

7.0h

Cost

$0.84

Quality

88%

Fine-tune this model

#27QLoRA

DeepSeek R1 Distill 14B

DeepSeek R1 · 14.8B

GPU

Time

13h

Cost

$1.6

Quality

88%

Fine-tune this model

#28ReFT

DeepSeek R1 Distill 32B

DeepSeek R1 · 32.8B

GPU

H100 SXM

Time

1.9h

Cost

$3.6

Quality

83%

Fine-tune this model

#29QLoRA

DeepSeek R1 Distill 70B

DeepSeek R1 · 70.6B

GPU

H100 SXM

Time

4.1h

Cost

$7.7

Quality

88%

Fine-tune this model

#30QLoRA

DeepSeek R1

DeepSeek R1 · 671B MoE (37B active)

GPU

T4x31

Time

24h

Cost

$91

Quality

88%

Fine-tune this model

#31QLoRA

DeepSeek V3

DeepSeek V3 · 671B MoE (37B active)

GPU

T4x31

Time

24h

Cost

$91

Quality

88%

Fine-tune this model

#32QLoRA

Llama 3 70B 1M Context

Llama 3 · 70.6B

GPU

H100 SXM

Time

4.1h

Cost

$7.7

Quality

88%

Fine-tune this model

#33QLoRA

Llama 3 70B

Llama 3 · 70.6B

GPU

H100 SXM

Time

4.1h

Cost

$7.7

Quality

88%

Fine-tune this model

#34QLoRA

Mixtral 8x7B

Mixtral · 46.7B MoE (12.9B active)

GPU

H100 SXM

Time

2.7h

Cost

$5.1

Quality

88%

Fine-tune this model

#35QLoRA

Mixtral 8x7B Instruct

Mixtral · 46.7B MoE (12.9B active)

GPU

H100 SXM

Time

2.7h

Cost

$5.1

Quality

88%

Fine-tune this model

#36QLoRA

Gemma 2 9B

Gemma 2 · 9.2B

GPU

Time

8.1h

Cost

$0.97

Quality

88%

Fine-tune this model

#37ReFT

Phi 4 Mini

Phi · 3.8B

GPU

Time

3.3h

Cost

$0.40

Quality

83%

Fine-tune this model

#38QLoRA

Phi-4

Phi · 14.7B

GPU

Time

13h

Cost

$1.5

Quality

88%

Fine-tune this model

#39QLoRA

Llama 2 7B

Llama 2 · 7B

GPU

Time

6.1h

Cost

$0.74

Quality

88%

Fine-tune this model

#40QLoRA

Llama 2 13B

Llama 2 · 13B

GPU

Time

11h

Cost

$1.4

Quality

88%

Fine-tune this model

#41QLoRA

Llama 2 70B

Llama 2 · 70B

GPU

H100 SXM

Time

4.0h

Cost

$7.6

Quality

88%

Fine-tune this model

#42NEFTune

Llama Guard 3 1B

Llama Guard · 1B

GPU

Time

53m

Cost

$0.11

Quality

85%

Fine-tune this model

#43QLoRA

Code Llama 7B

Code Llama · 7B

GPU

Time

6.1h

Cost

$0.74

Quality

88%

Fine-tune this model

#44QLoRA

Llama Guard 3 8B

Llama Guard · 8B

GPU

Time

7.0h

Cost

$0.84

Quality

88%

Fine-tune this model

#45QLoRA

Llama 3.3 8B

Llama 3.3 · 8B

GPU

Time

7.0h

Cost

$0.84

Quality

88%

Fine-tune this model

#46QLoRA

Llama 3.2 11B Vision

Llama 3.2 · 11B

GPU

Time

9.6h

Cost

$1.2

Quality

88%

Fine-tune this model

#47QLoRA

Code Llama 13B

Code Llama · 13B

GPU

Time

11h

Cost

$1.4

Quality

88%

Fine-tune this model

#48ReFT

Code Llama 34B

Code Llama · 34B

GPU

H100 SXM

Time

2.0h

Cost

$3.7

Quality

83%

Fine-tune this model

#49QLoRA

Llama 3.1 Nemotron 51B

Llama 3.1 · 51B

GPU

H100 SXM

Time

2.9h

Cost

$5.5

Quality

88%

Fine-tune this model

#50QLoRA

Code Llama 70B

Code Llama · 70B

GPU

H100 SXM

Time

4.0h

Cost

$7.6

Quality

88%

Fine-tune this model

#51QLoRA

Llama 3.3 70B

Llama 3.3 · 70.6B

GPU

H100 SXM

Time

4.1h

Cost

$7.7

Quality

88%

Fine-tune this model

#52QLoRA

Llama 3.1 Nemotron 70B Instruct

Llama 3.1 · 70.6B

GPU

H100 SXM

Time

4.1h

Cost

$7.7

Quality

88%

Fine-tune this model

#53QLoRA

Llama 3.1 Nemotron 70B Reward

Llama 3.1 · 70.6B

GPU

H100 SXM

Time

4.1h

Cost

$7.7

Quality

88%

Fine-tune this model

#54QLoRA

Llama 3.2 90B Vision Instruct

Llama 3.2 · 88.8B

GPU

H100 SXM

Time

5.1h

Cost

$9.6

Quality

88%

Fine-tune this model

#55QLoRA

Llama 3.2 90B Vision

Llama 3.2 · 90B

GPU

H100 SXM

Time

5.2h

Cost

$9.8

Quality

88%

Fine-tune this model

#56QLoRA

Llama 4 Scout

Llama 4 · 109B MoE (17B active)

GPU

B200 SXM

Time

2.8h

Cost

$12

Quality

88%

Fine-tune this model

#57QLoRA

Llama 4 Maverick

Llama 4 · 400B MoE (17B active)

GPU

B200 NVL (pair)

Time

5.1h

Cost

$53

Quality

88%

Fine-tune this model

#58QLoRA

Llama 3.1 405B

Llama 3.1 · 405B

GPU

B200 NVL (pair)

Time

5.1h

Cost

$54

Quality

88%

Fine-tune this model

#59NEFTune

Llama 4 Behemoth

Llama 4 · 2000B MoE (400B active)

GPU

T4x1645

Time

1.4h

Cost

$270

Quality

85%

Fine-tune this model

#60NEFTune

Qwen 2.5 0.5B

Qwen 2.5 · 500M

GPU

Time

26m

Cost

$0.05

Quality

85%

Fine-tune this model

#61NEFTune

Qwen 3 0.6B

Qwen 3 · 600M

GPU

Time

32m

Cost

$0.06

Quality

85%

Fine-tune this model

#62GaLore

Qwen 2.5 Coder 1.5B

Qwen 2.5 Coder · 1.5B

GPU

Time

1.3h

Cost

$0.16

Quality

97%

Fine-tune this model

#63GaLore

Qwen 3 1.7B

Qwen 3 · 1.7B

GPU

Time

1.5h

Cost

$0.18

Quality

97%

Fine-tune this model

#64GaLore

Qwen 2 VL 2B

Qwen 2 VL · 2.2B

GPU

Time

1.9h

Cost

$0.23

Quality

97%

Fine-tune this model

#65GaLore

Qwen 2.5 Coder 3B

Qwen 2.5 Coder · 3.1B

GPU

Time

2.7h

Cost

$0.33

Quality

97%

Fine-tune this model

#66QLoRA

GTE Qwen2 7B

GTE · 7.6B

GPU

Time

6.7h

Cost

$0.80

Quality

88%

Fine-tune this model

#67QLoRA

Qwen 2 Audio 7B

Qwen 2 · 7.6B

GPU

Time

6.7h

Cost

$0.80

Quality

88%

Fine-tune this model

#68QLoRA

Qwen 2.5 Coder 7B

Qwen 2.5 Coder · 7.6B

GPU

Time

6.7h

Cost

$0.80

Quality

88%

Fine-tune this model

#69QLoRA

Qwen 2.5 Math 7B

Qwen 2.5 Math · 7.6B

GPU

Time

6.7h

Cost

$0.80

Quality

88%

Fine-tune this model

#70QLoRA

Qwen 2.5 VL 7B

Qwen 2.5 VL · 7.6B

GPU

Time

6.7h

Cost

$0.80

Quality

88%

Fine-tune this model

#71QLoRA

Qwen 1.5 MoE A2.7B

Qwen 1.5 · 14.3B MoE (2.7B active)

GPU

Time

13h

Cost

$1.5

Quality

88%

Fine-tune this model

#72QLoRA

Qwen 2.5 Coder 14B

Qwen 2.5 Coder · 14.7B

GPU

Time

13h

Cost

$1.5

Quality

88%

Fine-tune this model

#73ReFT

Qwen 3 30B-A3B

Qwen 3 · 30.5B MoE (3.3B active)

GPU

H100 SXM

Time

1.8h

Cost

$3.3

Quality

83%

Fine-tune this model

#74ReFT

Qwen 2.5 Coder 32B

Qwen 2.5 · 32.5B

GPU

H100 SXM

Time

1.9h

Cost

$3.5

Quality

83%

Fine-tune this model

#75QLoRA

Qwen 2.5 Math 72B

Qwen 2.5 Math · 72.7B

GPU

H100 SXM

Time

4.2h

Cost

$7.9

Quality

88%

Fine-tune this model

#76QLoRA

Qwen 2.5 VL 72B

Qwen 2.5 VL · 72.7B

GPU

H100 SXM

Time

4.2h

Cost

$7.9

Quality

88%

Fine-tune this model

#77QLoRA

Qwen 3 235B

Qwen 3 · 235B MoE (22B active)

GPU

B200 SXM

Time

5.9h

Cost

$27

Quality

88%

Fine-tune this model

#78QLoRA

Codestral Mamba 7B

Codestral · 7.3B

GPU

Time

6.4h

Cost

$0.77

Quality

88%

Fine-tune this model

#79QLoRA

Ministral 8B

Ministral · 8B

GPU

Time

7.0h

Cost

$0.84

Quality

88%

Fine-tune this model

#80QLoRA

Mistral Nemo 12B

Mistral Nemo · 12B

GPU

Time

11h

Cost

$1.3

Quality

88%

Fine-tune this model

#81QLoRA

Pixtral 12B

Pixtral · 12B

GPU

Time

11h

Cost

$1.3

Quality

88%

Fine-tune this model

#82ReFT

Codestral 22B

Codestral · 22B

GPU

H100 SXM

Time

1.3h

Cost

$2.4

Quality

83%

Fine-tune this model

#83ReFT

Mistral Small 24B

Mistral Small · 24B

GPU

H100 SXM

Time

1.4h

Cost

$2.6

Quality

83%

Fine-tune this model

#84ReFT

Mistral Small 3.1 24B

Mistral Small · 24B

GPU

H100 SXM

Time

1.4h

Cost

$2.6

Quality

83%

Fine-tune this model

#85QLoRA

Mistral Medium 3

Mistral · 70B

GPU

H100 SXM

Time

4.0h

Cost

$7.6

Quality

88%

Fine-tune this model

#86QLoRA

Mistral Large 2411

Mistral Large · 123B

GPU

B200 SXM

Time

3.1h

Cost

$14

Quality

88%

Fine-tune this model

#87QLoRA

Mistral Large 2

Mistral Large · 123B

GPU

B200 SXM

Time

3.1h

Cost

$14

Quality

88%

Fine-tune this model

#88QLoRA

Mixtral 8x22B

Mixtral · 141B MoE (39B active)

GPU

B200 SXM

Time

3.6h

Cost

$16

Quality

88%

Fine-tune this model

#89ReFT

DeepSeek Coder 6.7B

DeepSeek Coder · 6.7B

GPU

Time

5.9h

Cost

$0.70

Quality

83%

Fine-tune this model

#90QLoRA

DeepSeek Math 7B

DeepSeek Math · 7.2B

GPU

Time

6.3h

Cost

$0.76

Quality

88%

Fine-tune this model

#91QLoRA

DeepSeek V2 Lite

DeepSeek V2 · 15.7B MoE (2.4B active)

GPU

Time

14h

Cost

$1.6

Quality

88%

Fine-tune this model

#92QLoRA

DeepSeek MoE 16B

DeepSeek MoE · 16.4B MoE (2.8B active)

GPU

Time

14h

Cost

$1.7

Quality

88%

Fine-tune this model

#93ReFT

DeepSeek Coder 33B

DeepSeek Coder · 33B

GPU

H100 SXM

Time

1.9h

Cost

$3.6

Quality

83%

Fine-tune this model

#94QLoRA

DeepSeek LLM 67B

DeepSeek LLM · 67B

GPU

H100 SXM

Time

3.9h

Cost

$7.3

Quality

88%

Fine-tune this model

#95QLoRA

DeepSeek Coder V2 236B

DeepSeek Coder V2 · 236B MoE (21B active)

GPU

B200 SXM

Time

6.0h

Cost

$27

Quality

88%

Fine-tune this model

#96QLoRA

DeepSeek V2.5

DeepSeek V2 · 236B MoE (21B active)

GPU

B200 SXM

Time

6.0h

Cost

$27

Quality

88%

Fine-tune this model

#97GaLore

Phi 1.5

Phi · 1.3B

GPU

Time

1.1h

Cost

$0.14

Quality

97%

Fine-tune this model

#98GaLore

Phi 1

Phi · 1.3B

GPU

Time

1.1h

Cost

$0.14

Quality

97%

Fine-tune this model

#99GaLore

Gemma 3 2B

Gemma 3 · 2B

GPU

Time

1.8h

Cost

$0.21

Quality

97%

Fine-tune this model

#100GaLore

Gemma 1.1 2B

Gemma · 2.5B

GPU

Time

2.2h

Cost

$0.26

Quality

97%

Fine-tune this model

#101GaLore

Gemma 2 2B

Gemma 2 · 2.6B

GPU

Time

2.3h

Cost

$0.27

Quality

97%

Fine-tune this model

#102GaLore

RecurrentGemma 2B

RecurrentGemma · 2.7B

GPU

Time

2.4h

Cost

$0.28

Quality

97%

Fine-tune this model

#103GaLore

PaLI-Gemma 3B

PaLI-Gemma · 2.9B

GPU

Time

2.5h

Cost

$0.30

Quality

97%

Fine-tune this model

#104ReFT

Phi 3 Mini 3.8B

Phi 3 · 3.8B

GPU

Time

3.3h

Cost

$0.40

Quality

83%

Fine-tune this model

#105ReFT

Phi 3.5 Vision

Phi 3.5 · 4.2B

GPU

Time

3.7h

Cost

$0.44

Quality

83%

Fine-tune this model

#106ReFT

Gemma 3 4B

Gemma 3 · 4.3B

GPU

Time

3.8h

Cost

$0.45

Quality

83%

Fine-tune this model

#107QLoRA

Phi 3 Small 7B

Phi 3 · 7B

GPU

Time

6.1h

Cost

$0.74

Quality

88%

Fine-tune this model

#108QLoRA

CodeGemma 7B

Gemma · 8.5B

GPU

Time

7.4h

Cost

$0.89

Quality

88%

Fine-tune this model

#109QLoRA

Gemma 3 12B

Gemma 3 · 12B

GPU

Time

11h

Cost

$1.3

Quality

88%

Fine-tune this model

#110QLoRA

Phi 3 Medium 14B

Phi 3 · 14B

GPU

Time

12h

Cost

$1.5

Quality

88%

Fine-tune this model

#111ReFT

Gemma 2 27B

Gemma 2 · 27B

GPU

H100 SXM

Time

1.6h

Cost

$2.9

Quality

83%

Fine-tune this model

#112ReFT

Gemma 3 27B

Gemma 3 · 27B

GPU

H100 SXM

Time

1.6h

Cost

$2.9

Quality

83%

Fine-tune this model

#113QLoRA

Phi 3.5 MoE

Phi · 41.9B MoE (6.6B active)

GPU

H100 SXM

Time

2.4h

Cost

$4.6

Quality

88%

Fine-tune this model

#114QLoRA

Dolphin 2.9 72B

Dolphin · 72B

GPU

H100 SXM

Time

4.1h

Cost

$7.8

Quality

88%

Fine-tune this model

#115NEFTune

BGE Base EN v1.5

BGE · 110M

GPU

Time

Cost

$0.01

Quality

85%

Fine-tune this model

#116NEFTune

SmolLM 135M

SmolLM · 135M

GPU

Time

Cost

$0.01

Quality

85%

Fine-tune this model

#117NEFTune

Nomic Embed Text v1.5

Nomic Embed · 137M

GPU

Time

Cost

$0.01

Quality

85%

Fine-tune this model

#118NEFTune

Whisper Small

Whisper · 244M

GPU

Time

13m

Cost

$0.03

Quality

85%

Fine-tune this model

#119NEFTune

NV EmbedQA E5 v5

NV EmbedQA · 330M

GPU

Time

17m

Cost

$0.03

Quality

85%

Fine-tune this model

#120NEFTune

NV Retriever v1

NV Retriever · 330M

GPU

Time

17m

Cost

$0.03

Quality

85%

Fine-tune this model

#121NEFTune

BGE Large EN v1.5

BGE · 335M

GPU

Time

18m

Cost

$0.04

Quality

85%

Fine-tune this model

#122NEFTune

SmolLM 360M

SmolLM · 360M

GPU

Time

19m

Cost

$0.04

Quality

85%

Fine-tune this model

#123NEFTune

H2O Danube3 500M

H2O Danube · 500M

GPU

Time

26m

Cost

$0.05

Quality

85%

Fine-tune this model

#124NEFTune

Cohere Embed English v3

Embed · 500M

GPU

Time

26m

Cost

$0.05

Quality

85%

Fine-tune this model

#125NEFTune

InfoXLM Large

InfoXLM · 550M

GPU

Time

29m

Cost

$0.06

Quality

85%

Fine-tune this model

#126NEFTune

Multilingual E5 Large

E5 · 560M

GPU

Time

29m

Cost

$0.06

Quality

85%

Fine-tune this model

#127NEFTune

BGE M3

BGE · 568M

GPU

Time

30m

Cost

$0.06

Quality

85%

Fine-tune this model

#128NEFTune

Jina Embeddings v3

Jina Embeddings · 570M

GPU

Time

30m

Cost

$0.06

Quality

85%

Fine-tune this model

#129NEFTune

Parakeet CTC 0.6B

Parakeet · 600M

GPU

Time

32m

Cost

$0.06

Quality

85%

Fine-tune this model

#130NEFTune

Whisper Medium

Whisper · 769M

GPU

Time

40m

Cost

$0.08

Quality

85%

Fine-tune this model

#131NEFTune

Florence 2 Large

Florence · 770M

GPU

Time

40m

Cost

$0.08

Quality

85%

Fine-tune this model

#132NEFTune

Canary 1B

Canary · 1B

GPU

Time

53m

Cost

$0.11

Quality

85%

Fine-tune this model

#133NEFTune

Parakeet TDT 1.1B

Parakeet · 1.1B

GPU

Time

58m

Cost

$0.12

Quality

85%

Fine-tune this model

#134NEFTune

SantaCoder 1.1B

SantaCoder · 1.1B

GPU

Time

58m

Cost

$0.12

Quality

85%

Fine-tune this model

#135GaLore

Whisper Large V3

Whisper · 1.6B

GPU

Time

1.4h

Cost

$0.16

Quality

97%

Fine-tune this model

#136GaLore

SmolLM2 1.7B

SmolLM2 · 1.7B

GPU

Time

1.5h

Cost

$0.18

Quality

97%

Fine-tune this model

#137GaLore

SeamlessM4T v2 Large

SeamlessM4T · 2.3B

GPU

Time

2.0h

Cost

$0.24

Quality

97%

Fine-tune this model

#138GaLore

OpenELM 3B

OpenELM · 3B

GPU

Time

2.6h

Cost

$0.32

Quality

97%

Fine-tune this model

#139GaLore

VILA 1.5 3B

VILA · 3B

GPU

Time

2.6h

Cost

$0.32

Quality

97%

Fine-tune this model

#140GaLore

StableLM Zephyr 3B

StableLM · 3B

GPU

Time

2.6h

Cost

$0.32

Quality

97%

Fine-tune this model

#141GaLore

BTLM 3B

BTLM · 3B

GPU

Time

2.6h

Cost

$0.32

Quality

97%

Fine-tune this model

#142GaLore

StarCoder2 3B

StarCoder2 · 3B

GPU

Time

2.7h

Cost

$0.32

Quality

97%

Fine-tune this model

#143ReFT

Replit Code v1.5 3B

Replit Code · 3.3B

GPU

Time

2.9h

Cost

$0.35

Quality

83%

Fine-tune this model

#144ReFT

DALL-E 3

DALL-E · 3.5B

GPU

Time

3.1h

Cost

$0.37

Quality

83%

Fine-tune this model

#145ReFT

Stable Diffusion XL 1.0

Stable Diffusion · 3.5B

GPU

Time

3.1h

Cost

$0.37

Quality

83%

Fine-tune this model

#146ReFT

Minitron 4B

Nemotron · 4B

GPU

Time

3.5h

Cost

$0.42

Quality

83%

Fine-tune this model

#147ReFT

Nemotron Mini 4B

Nemotron · 4B

GPU

Time

3.5h

Cost

$0.42

Quality

83%

Fine-tune this model

#148ReFT

ChatGLM3 6B

ChatGLM3 · 6B

GPU

Time

5.3h

Cost

$0.63

Quality

83%

Fine-tune this model

#149ReFT

Yi 6B 200K

Yi · 6B

GPU

Time

5.3h

Cost

$0.63

Quality

83%

Fine-tune this model

#150ReFT

SciGLM 6B

SciGLM · 6.2B

GPU

Time

5.4h

Cost

$0.65

Quality

83%

Fine-tune this model

#151ReFT

MPT 7B

MPT · 6.7B

GPU

Time

5.9h

Cost

$0.70

Quality

83%

Fine-tune this model

#152ReFT

StarCoder2 7B

StarCoder2 · 6.7B

GPU

Time

5.9h

Cost

$0.71

Quality

83%

Fine-tune this model

#153QLoRA

OLMo 2 7B

OLMo 2 · 7B

GPU

Time

6.1h

Cost

$0.74

Quality

88%

Fine-tune this model

#154QLoRA

Command R 7B

Command R · 7B

GPU

Time

6.1h

Cost

$0.74

Quality

88%

Fine-tune this model

#155QLoRA

Falcon 7B

Falcon · 7B

GPU

Time

6.1h

Cost

$0.74

Quality

88%

Fine-tune this model

#156QLoRA

Zephyr 7B

Zephyr · 7B

GPU

Time

6.1h

Cost

$0.74

Quality

88%

Fine-tune this model

#157QLoRA

Vicuna 7B

Vicuna · 7B

GPU

Time

6.1h

Cost

$0.74

Quality

88%

Fine-tune this model

#158QLoRA

OpenHermes 2.5 7B

OpenHermes · 7B

GPU

Time

6.1h

Cost

$0.74

Quality

88%

Fine-tune this model

#159QLoRA

Baichuan 2 7B

Baichuan 2 · 7B

GPU

Time

6.1h

Cost

$0.74

Quality

88%

Fine-tune this model

#160QLoRA

SaulLM 7B

SaulLM · 7.2B

GPU

Time

6.3h

Cost

$0.76

Quality

88%

Fine-tune this model

#161QLoRA

FinGPT 7B

FinGPT · 7.2B

GPU

Time

6.3h

Cost

$0.76

Quality

88%

Fine-tune this model

#162QLoRA

Prometheus 2 7B

Prometheus · 7.2B

GPU

Time

6.3h

Cost

$0.76

Quality

88%

Fine-tune this model

#163QLoRA

Falcon Mamba 7B

Falcon Mamba · 7.3B

GPU

Time

6.4h

Cost

$0.76

Quality

88%

Fine-tune this model

#164QLoRA

Marco O1

Marco · 7.6B

GPU

Time

6.7h

Cost

$0.80

Quality

88%

Fine-tune this model

#165QLoRA

InternLM 2.5 7B

InternLM 2.5 · 7.7B

GPU

Time

6.8h

Cost

$0.81

Quality

88%

Fine-tune this model

#166QLoRA

NV Embed v2

NV Embed · 7.8B

GPU

Time

6.9h

Cost

$0.82

Quality

88%

Fine-tune this model

#167QLoRA

Aya 23 8B

Aya · 8B

GPU

Time

7.0h

Cost

$0.84

Quality

88%

Fine-tune this model

#168QLoRA

Minitron 8B

Nemotron · 8B

GPU

Time

7.0h

Cost

$0.84

Quality

88%

Fine-tune this model

#169QLoRA

GPT-4o Mini

GPT-4 · 8B

GPU

Time

7.0h

Cost

$0.84

Quality

88%

Fine-tune this model

#170QLoRA

Hermes 3 8B

Hermes 3 · 8B

GPU

Time

7.0h

Cost

$0.84

Quality

88%

Fine-tune this model

#171QLoRA

Yi Coder 9B

Yi Coder · 8.8B

GPU

Time

7.7h

Cost

$0.92

Quality

88%

Fine-tune this model

#172QLoRA

Yi 1.5 9B

Yi 1.5 · 8.8B

GPU

Time

7.7h

Cost

$0.93

Quality

88%

Fine-tune this model

#173QLoRA

GLM-4 9B

GLM-4 · 9.4B

GPU

Time

8.2h

Cost

$0.99

Quality

88%

Fine-tune this model

#174QLoRA

ChatGLM4 9B

ChatGLM · 9.4B

GPU

Time

8.2h

Cost

$0.99

Quality

88%

Fine-tune this model

#175QLoRA

SOLAR 10.7B

SOLAR · 10.7B

GPU

Time

9.4h

Cost

$1.1

Quality

88%

Fine-tune this model

#176QLoRA

Falcon 11B

Falcon · 11B

GPU

Time

9.6h

Cost

$1.2

Quality

88%

Fine-tune this model

#177QLoRA

Amazon Nova Lite

Nova · 12B

GPU

Time

11h

Cost

$1.3

Quality

88%

Fine-tune this model

#178QLoRA

FLUX.1 Dev

FLUX · 12B

GPU

Time

11h

Cost

$1.3

Quality

88%

Fine-tune this model

#179QLoRA

StableLM 2 12B

StableLM 2 · 12.1B

GPU

Time

11h

Cost

$1.3

Quality

88%

Fine-tune this model

#180QLoRA

KULLM 12.8B

KULLM · 12.8B

GPU

Time

11h

Cost

$1.3

Quality

88%

Fine-tune this model

#181QLoRA

OLMo 2 13B

OLMo 2 · 13B

GPU

Time

11h

Cost

$1.4

Quality

88%

Fine-tune this model

#182QLoRA

Baichuan 2 13B

Baichuan 2 · 13B

GPU

Time

11h

Cost

$1.4

Quality

88%

Fine-tune this model

#183QLoRA

Vicuna 13B

Vicuna · 13B

GPU

Time

11h

Cost

$1.4

Quality

88%

Fine-tune this model

#184QLoRA

Orca 2 13B

Orca · 13B

GPU

Time

11h

Cost

$1.4

Quality

88%

Fine-tune this model

#185QLoRA

VILA 1.5 13B

VILA · 13B

GPU

Time

11h

Cost

$1.4

Quality

88%

Fine-tune this model

#186QLoRA

ELYZA 13B

ELYZA · 13B

GPU

Time

11h

Cost

$1.4

Quality

88%

Fine-tune this model

#187QLoRA

Cerebras GPT 13B

Cerebras GPT · 13B

GPU

Time

11h

Cost

$1.4

Quality

88%

Fine-tune this model

#188QLoRA

Nekomata 14B

Nekomata · 14B

GPU

Time

12h

Cost

$1.5

Quality

88%

Fine-tune this model

#189QLoRA

RWKV-6 14B

RWKV · 14.1B

GPU

Time

12h

Cost

$1.5

Quality

88%

Fine-tune this model

#190QLoRA

Nemotron 15B

Nemotron · 15B

GPU

Time

13h

Cost

$1.6

Quality

88%

Fine-tune this model

#191QLoRA

OctoCoder 15B

OctoCoder · 15.5B

GPU

Time

14h

Cost

$1.6

Quality

88%

Fine-tune this model

#192QLoRA

StarCoder2 15B

StarCoder2 · 15.5B

GPU

Time

14h

Cost

$1.6

Quality

88%

Fine-tune this model

#193QLoRA

CodeGen2 16B

CodeGen2 · 16B

GPU

Time

14h

Cost

$1.7

Quality

88%

Fine-tune this model

#194QLoRA

CogVLM2 19B

CogVLM2 · 19B

GPU

Time

17h

Cost

$2.0

Quality

88%

Fine-tune this model

#195QLoRA

InternLM 2.5 20B

InternLM 2.5 · 19.9B

GPU

Time

17h

Cost

$2.1

Quality

88%

Fine-tune this model

#196QLoRA

GigaChat 20B

GigaChat · 20B

GPU

Time

18h

Cost

$2.1

Quality

88%

Fine-tune this model

#197QLoRA

Claude 3.5 Haiku

Claude · 20B

GPU

Time

18h

Cost

$2.1

Quality

88%

Fine-tune this model

#198QLoRA

GPT-3.5 Turbo

GPT-3.5 · 20B

GPU

Time

18h

Cost

$2.1

Quality

88%

Fine-tune this model

#199QLoRA

InternLM 20B

InternLM · 20B

GPU

Time

18h

Cost

$2.1

Quality

88%

Fine-tune this model

#200ReFT

Solar Pro 22B

Solar · 22B

GPU

H100 SXM

Time

1.3h

Cost

$2.4

Quality

83%

Fine-tune this model

#201ReFT

InternVL2 26B

InternVL2 · 26B

GPU

H100 SXM

Time

1.5h

Cost

$2.8

Quality

83%

Fine-tune this model

#202ReFT

JAIS 30B

JAIS · 30B

GPU

H100 SXM

Time

1.7h

Cost

$3.3

Quality

83%

Fine-tune this model

#203ReFT

MPT 30B

MPT · 30B

GPU

H100 SXM

Time

1.7h

Cost

$3.3

Quality

83%

Fine-tune this model

#204ReFT

Vicuna 33B

Vicuna · 33B

GPU

H100 SXM

Time

1.9h

Cost

$3.6

Quality

83%

Fine-tune this model

#205ReFT

WizardCoder 33B

WizardCoder · 33B

GPU

H100 SXM

Time

1.9h

Cost

$3.6

Quality

83%

Fine-tune this model

#206ReFT

Yi 1.5 34B

Yi 1.5 · 34.4B

GPU

H100 SXM

Time

2.0h

Cost

$3.7

Quality

83%

Fine-tune this model

#207QLoRA

Aya 23 35B

Aya · 35B

GPU

H100 SXM

Time

2.0h

Cost

$3.8

Quality

88%

Fine-tune this model

#208QLoRA

Command R

Command R · 35B

GPU

H100 SXM

Time

2.0h

Cost

$3.8

Quality

88%

Fine-tune this model

#209QLoRA

Command R (August 2024)

Command R · 35B

GPU

H100 SXM

Time

2.0h

Cost

$3.8

Quality

88%

Fine-tune this model

#210QLoRA

Falcon 40B

Falcon · 40B

GPU

H100 SXM

Time

2.3h

Cost

$4.3

Quality

88%

Fine-tune this model

#211QLoRA

VILA 1.5 40B

VILA · 40B

GPU

H100 SXM

Time

2.3h

Cost

$4.3

Quality

88%

Fine-tune this model

#212QLoRA

Amazon Nova Pro

Nova · 50B

GPU

H100 SXM

Time

2.9h

Cost

$5.4

Quality

88%

Fine-tune this model

#213QLoRA

Gemini 2.0 Flash

Gemini · 50B MoE (15B active)

GPU

H100 SXM

Time

2.9h

Cost

$5.4

Quality

88%

Fine-tune this model

#214QLoRA

Gemini 1.5 Flash

Gemini · 50B MoE (12B active)

GPU

H100 SXM

Time

2.9h

Cost

$5.4

Quality

88%

Fine-tune this model

#215QLoRA

Jamba 1.5 Mini

Jamba · 52B

GPU

H100 SXM

Time

3.0h

Cost

$5.6

Quality

88%

Fine-tune this model

#216QLoRA

Jamba Instruct

Jamba · 52B MoE (12B active)

GPU

H100 SXM

Time

3.0h

Cost

$5.6

Quality

88%

Fine-tune this model

#217QLoRA

WizardMath 70B

WizardMath · 70B

GPU

H100 SXM

Time

4.0h

Cost

$7.6

Quality

88%

Fine-tune this model

#218QLoRA

Claude Sonnet 4

Claude · 70B

GPU

H100 SXM

Time

4.0h

Cost

$7.6

Quality

88%

Fine-tune this model

#219QLoRA

o1-mini

o1 · 70B

GPU

H100 SXM

Time

4.0h

Cost

$7.6

Quality

88%

Fine-tune this model

#220QLoRA

o3-mini

o3 · 70B

GPU

H100 SXM

Time

4.0h

Cost

$7.6

Quality

88%

Fine-tune this model

#221QLoRA

Claude 3 Sonnet

Claude · 70B

GPU

H100 SXM

Time

4.0h

Cost

$7.6

Quality

88%

Fine-tune this model

#222QLoRA

Reka Core

Reka · 70B

GPU

H100 SXM

Time

4.0h

Cost

$7.6

Quality

88%

Fine-tune this model

#223QLoRA

Meditron 70B

Meditron · 70B

GPU

H100 SXM

Time

4.0h

Cost

$7.6

Quality

88%

Fine-tune this model

#224QLoRA

Japanese StableLM 70B

StableLM · 70B

GPU

H100 SXM

Time

4.0h

Cost

$7.6

Quality

88%

Fine-tune this model

#225QLoRA

Hermes 3 70B

Hermes 3 · 70.6B

GPU

H100 SXM

Time

4.1h

Cost

$7.7

Quality

88%

Fine-tune this model

#226QLoRA

Nemotron 70B

Nemotron · 70.6B

GPU

H100 SXM

Time

4.1h

Cost

$7.7

Quality

88%

Fine-tune this model

#227QLoRA

Inflection 3

Inflection · 100B

GPU

H100 SXM

Time

5.7h

Cost

$11

Quality

88%

Fine-tune this model

#228QLoRA

YaLM 100B

YaLM · 100B

GPU

H100 SXM

Time

5.7h

Cost

$11

Quality

88%

Fine-tune this model

#229QLoRA

Yi-Large

Yi · 103B MoE (24B active)

GPU

H100 SXM

Time

5.9h

Cost

$11

Quality

88%

Fine-tune this model

#230QLoRA

Command R+

Command R · 104B

GPU

H100 SXM

Time

6.0h

Cost

$11

Quality

88%

Fine-tune this model

#231QLoRA

Command A

Command · 111B

GPU

B200 SXM

Time

2.8h

Cost

$13

Quality

88%

Fine-tune this model

#232QLoRA

DBRX Base

DBRX · 132B MoE (36B active)

GPU

B200 SXM

Time

3.3h

Cost

$15

Quality

88%

Fine-tune this model

#233QLoRA

DBRX Instruct

DBRX · 132B MoE (36B active)

GPU

B200 SXM

Time

3.3h

Cost

$15

Quality

88%

Fine-tune this model

#234QLoRA

Gemini 1.5 Pro

Gemini · 175B MoE (40B active)

GPU

B200 SXM

Time

4.4h

Cost

$20

Quality

88%

Fine-tune this model

#235QLoRA

Claude 3 Opus

Claude · 175B

GPU

B200 SXM

Time

4.4h

Cost

$20

Quality

88%

Fine-tune this model

#236QLoRA

Falcon 180B

Falcon · 180B

GPU

B200 SXM

Time

4.6h

Cost

$20

Quality

88%

Fine-tune this model

#237QLoRA

Claude Opus 4

Claude · 200B

GPU

B200 SXM

Time

5.1h

Cost

$23

Quality

88%

Fine-tune this model

#238QLoRA

GPT-4o

GPT-4 · 200B MoE (50B active)

GPU

B200 SXM

Time

5.1h

Cost

$23

Quality

88%

Fine-tune this model

#239QLoRA

GPT-4 Turbo

GPT-4 · 200B MoE (50B active)

GPU

B200 SXM

Time

5.1h

Cost

$23

Quality

88%

Fine-tune this model

#240QLoRA

o1

o1 · 200B MoE (50B active)

GPU

B200 SXM

Time

5.1h

Cost

$23

Quality

88%

Fine-tune this model

#241QLoRA

Grok-2

Grok · 314B MoE (50B active)

GPU

B200 NVL (pair)

Time

4.0h

Cost

$42

Quality

88%

Fine-tune this model

#242QLoRA

Nemotron 340B

Nemotron · 340B

GPU

B200 NVL (pair)

Time

4.3h

Cost

$45

Quality

88%

Fine-tune this model

#243QLoRA

Snowflake Arctic 128x3B

Arctic · 395B MoE (17B active)

GPU

B200 NVL (pair)

Time

5.0h

Cost

$52

Quality

88%

Fine-tune this model

#244QLoRA

Jamba 1.5 Large

Jamba · 398B

GPU

B200 NVL (pair)

Time

5.0h

Cost

$53

Quality

88%

Fine-tune this model

#245QLoRA

Snowflake Arctic 480B

Arctic · 480B MoE (17B active)

GPU

B200 NVL (pair)

Time

6.1h

Cost

$64

Quality

88%

Fine-tune this model

#246DoRA

Megatron-Turing NLG 530B

Megatron-Turing · 530B

GPU

T4x81

Time

7.4h

Cost

$72

Quality

94%

Fine-tune this model

#247NEFTune

Gemini 2.0 Pro

Gemini · 600B MoE (150B active)

GPU

T4x494

Time

1.4h

Cost

$81

Quality

85%

Fine-tune this model

#248NEFTune

Grok 3

Grok · 600B MoE (120B active)

GPU

T4x494

Time

1.4h

Cost

$81

Quality

85%

Fine-tune this model

#249ReFT

Kimi K2.5

Kimi · 1000B MoE (32B active)

GPU

T4x145

Time

7.8h

Cost

$135

Quality

83%

Fine-tune this model

#250NEFTune

GPT-4.5 Preview

GPT · 1500B MoE (300B active)

GPU

T4x1234

Time

1.4h

Cost

$203

Quality

85%

Fine-tune this model

Showing 250 of 250 models

Ready to fine-tune your model?

Use the training calculator to get detailed cost estimates for any model, method, and GPU combination.

Open Training Calculator