Local AI models, ranked for phones

All 93 models we track, scored on the Galaxy S26 Ultra (16GB) as the reference flagship. The score blends fit headroom and decode speed — pick a model to see how it does on all 140 phones.

#		Model	Params	Quant · Size	Speed on Galaxy S26 Ultra	Tags	Verdict
1	S	Qwen3 0.6B	0.6B	Q8_0 · 0.6 GB	~63.6 tokens/s	chat	✓ Runs great
2	S	Qwen3 1.7B	1.7B	Q8_0 · 1.8 GB	~21.2 tokens/s	chat	✓ Runs great
3	S	Qwen3 4B	4B	Q4_K_M · 2.5 GB	~15.3 tokens/s	chat · coding	✓ Runs great
4	S	Llama 3.2 1B	1.2B	Q4_K_M · 0.8 GB	~47.7 tokens/s	chat	✓ Runs great
5	S	Llama 3.2 3B	3.2B	Q4_K_M · 2 GB	~19.1 tokens/s	chat	✓ Runs great
6	S	Gemma 3 1B	1B	Q4_K_M · 0.8 GB	~47.7 tokens/s	chat	✓ Runs great
7	S	Gemma 3 4B	4.3B	Q4_K_M · 2.5 GB	~15.3 tokens/s	chat · vision	✓ Runs great
8	S	Phi-4 Mini 3.8B	3.8B	Q4_K_M · 2.5 GB	~15.3 tokens/s	chat · reasoning	✓ Runs great
9	S	DeepSeek R1 Distill 1.5B	1.8B	Q4_K_M · 1.1 GB	~34.7 tokens/s	reasoning	✓ Runs great
10	S	SmolLM2 1.7B	1.7B	Q4_K_M · 1.1 GB	~34.7 tokens/s	chat	✓ Runs great
11	S	SmolLM3 3B	3.1B	Q4_K_M · 1.9 GB	~20.1 tokens/s	chat	✓ Runs great
12	S	Qwen 3.5 2B	2B	Q4_K_M · 1.3 GB	~29.4 tokens/s	chat · vision	✓ Runs great
13	S	Ternary Bonsai 8B	8B	Q2_0 · 2.2 GB	~17.3 tokens/s	chat · reasoning	✓ Runs great
14	S	Ternary Bonsai 4B	4B	Q2_0 · 1.1 GB	~34.7 tokens/s	chat	✓ Runs great
15	S	Ternary Bonsai 1.7B	1.7B	Q2_0 · 0.5 GB	~76.3 tokens/s	chat	✓ Runs great
16	S	Ministral 3 3B	3B	Q4_K_M · 2.1 GB	~18.2 tokens/s	chat · vision	✓ Runs great
17	S	LFM2.5 2.6B	2.7B	Q4_K_M · 1.7 GB	~22.4 tokens/s	chat · reasoning	✓ Runs great
18	S	LFM2.5 8B-A1B	8B	Q4_K_M · 5.2 GB	~58.7 tokens/s	chat	✓ Runs great
19	S	LFM2.5-VL 3B	3.1B	Q4_K_M · 1.7 GB	~22.4 tokens/s	chat · vision	✓ Runs great
20	S	OvisOCR2 0.8B	0.8B	Q4_K_M · 0.6 GB	~63.6 tokens/s	vision	✓ Runs great
21	S	G9v3 3B	3B	Q4_K_M · 1.9 GB	~20.1 tokens/s	chat · coding · reasoning	✓ Runs great
22	S	Tini Cybersec 8B-A1B	8.5B	Q4_K_M · 5.2 GB	~62.4 tokens/s	coding · reasoning	✓ Runs great
23	S	Ling 3.0 Tiny	7.9B	Q4_K_M · 4.9 GB	~43.9 tokens/s	chat · coding · reasoning	✓ Runs great
24	S	Granite 4.2 3B	3B	Q4_K_M · 2.2 GB	~17.3 tokens/s	chat · coding · reasoning	✓ Runs great
25	S	Qwen 3.5 4B	4B	Q4_K_M · 2.7 GB	~14.1 tokens/s	chat · vision · reasoning	✓ Runs great
26	S	Agents-A1 4B	4B	Q4_K_M · 2.7 GB	~14.1 tokens/s	chat · coding · vision · reasoning	✓ Runs great
27	S	Nanbeige 4.2 3B	4.2B	Q4_K_M · 2.7 GB	~14.1 tokens/s	chat · coding · reasoning	✓ Runs great
28	S	Nemotron 3 Nano 4B	4B	Q4_K_M · 2.8 GB	~13.6 tokens/s	chat · reasoning	✓ Runs great
29	S	AREX Turbo 4B	4.5B	Q4_K_M · 2.9 GB	~13.2 tokens/s	chat · reasoning	✓ Runs great
30	S	Fara 1.5 4B	4.5B	Q4_K_M · 2.9 GB	~13.2 tokens/s	vision · reasoning	✓ Runs great
31	S	Gemma 4 E2B	2B	Q4_K_M · 3.1 GB	~12.3 tokens/s	chat · vision	✓ Runs great
32	S	Bonsai 27B (1-bit)	27B	Q1_0 · 3.8 GB	~10 tokens/s	chat · coding · vision · reasoning	✓ Runs great
33	A	Mistral 7B v0.3	7.2B	Q4_K_M · 4.4 GB	~8.7 tokens/s	chat	✓ Runs great
34	A	DeepSeek R1 Distill 7B	7.6B	Q4_K_M · 4.7 GB	~8.1 tokens/s	reasoning	✓ Runs great
35	A	Llama 3.1 8B	8B	Q4_K_M · 4.9 GB	~7.8 tokens/s	chat · coding	! Runs, barely
36	A	Ministral 8B	8B	Q4_K_M · 4.9 GB	~7.8 tokens/s	chat	! Runs, barely
37	A	Qwen3 8B	8.2B	Q4_K_M · 5 GB	~7.6 tokens/s	chat · coding	! Runs, barely
38	A	Gemma 4 E4B	4B	Q4_K_M · 5 GB	~7.6 tokens/s	chat · vision	! Runs, barely
39	A	Ministral 3 8B	8B	Q4_K_M · 5.2 GB	~7.3 tokens/s	chat · vision	! Runs, barely
40	A	Granite 4.2 8B	8B	Q4_K_M · 5.3 GB	~7.2 tokens/s	chat · coding · reasoning	! Runs, barely
41	A	Qwen 3.5 9B	9B	Q4_K_M · 5.7 GB	~6.7 tokens/s	chat · vision · reasoning	! Runs, barely
42	A	Ornith 1.0 9B	9B	Q4_K_M · 5.6 GB	~6.8 tokens/s	coding · chat · reasoning	! Runs, barely
43	A	Qwythos 9B v2	9B	Q4_K_M · 5.7 GB	~6.7 tokens/s	chat · vision · reasoning	! Runs, barely
44	A	GRM 3.2 Cliff 9B	9.4B	Q4_K_M · 5.9 GB	~6.5 tokens/s	chat · coding · vision · reasoning	! Runs, barely
45	A	grug 9B (ProCreations)	9.4B	Q4_K_M · 5.9 GB	~6.5 tokens/s	chat · coding · vision · reasoning	! Runs, barely
46	A	Fara 1.5 9B	9.4B	Q4_K_M · 5.9 GB	~6.5 tokens/s	vision · reasoning	! Runs, barely
47	A	UI-Mate 9B	9B	Q4_K_M · 5.9 GB	~6.5 tokens/s	vision · reasoning	! Runs, barely
48	A	Gemma 4 12B	12B	Q4_0 · 7 GB	~5.5 tokens/s	chat · vision · reasoning	! Runs, barely
49	A	Gemma 3 12B	12.2B	Q4_K_M · 7.3 GB	~5.2 tokens/s	chat · vision	! Runs, barely
50	A	Ternary Bonsai 27B	27B	Q2_0 · 7.2 GB	~5.3 tokens/s	chat · coding · vision · reasoning	! Runs, barely
51	A	Grug 12B	12B	Q4_K_M · 7.7 GB	~5 tokens/s	chat · vision · reasoning	! Runs, barely
52	A	Ministral 3 14B	14B	Q4_K_M · 8.2 GB	~4.7 tokens/s	chat · vision · reasoning	! Runs, barely
53	B	Qwen3 14B	14.8B	Q4_K_M · 9 GB	~4.2 tokens/s	chat · coding · reasoning	! Runs, barely
54	B	Phi-4 14B	14.7B	Q4_K_M · 9.1 GB	~4.2 tokens/s	chat · reasoning	! Runs, barely
55	F	GPT-OSS 20B	21B	MXFP4 · 12.1 GB	—	chat · reasoning	✕ Won't fit
56	F	Gemma 4 26B-A4B	26B	Q4_0 · 14.4 GB	—	chat · vision · reasoning	✕ Won't fit
57	F	Qwen 3.6 27B	27B	Q4_K_M · 16.8 GB	—	chat · coding · vision · reasoning	✕ Won't fit
58	F	Qwen 3.5 27B	27B	Q4_K_M · 16.7 GB	—	chat · vision · reasoning · coding	✕ Won't fit
59	F	Muse Glimmer 30B	29.6B	K_QUANT_17GB · 16.8 GB	—	chat · coding · vision · reasoning	✕ Won't fit
60	F	Fara 1.5 27B	27B	Q4_K_M · 17.5 GB	—	vision · reasoning	✕ Won't fit
61	F	Qwen3.8 27B	27B	Q4_K_M · 19 GB	—	chat · coding · vision · reasoning	✕ Won't fit
62	F	UI-Mate 27B	27B	Q4_K_M · 17.5 GB	—	vision · reasoning	✕ Won't fit
63	F	Granite 4.2 30B	30B	Q4_K_M · 17.7 GB	—	chat · coding · reasoning	✕ Won't fit
64	F	Gemma 4 31B	31B	Q4_K_M · 18.3 GB	—	chat · vision · reasoning	✕ Won't fit
65	F	Qwen3 30B A3B	30.5B	Q4_K_M · 18.6 GB	—	chat · coding	✕ Won't fit
66	F	Salience 1.5 Flash	30B	Q4_K_M · 18.7 GB	—	chat · coding · vision · reasoning	✕ Won't fit
67	F	Qwen3 32B	32.8B	Q4_K_M · 19.8 GB	—	chat · coding · reasoning	✕ Won't fit
68	F	Qwen 3.5 35B-A3B	35B	Q4_K_M · 22 GB	—	chat · vision · reasoning	✕ Won't fit
69	F	Qwen 3.6 35B-A3B	35B	Q4_K_M · 22.1 GB	—	chat · coding · vision · reasoning	✕ Won't fit
70	F	Nemotron 3 Nano 30B-A3B	30B	Q4_K_M · 24.6 GB	—	chat · reasoning	✕ Won't fit
71	F	Nemotron 3 Nano Omni 30B-A3B	31B	Q4_K_M · 22.4 GB	—	chat · vision · reasoning	✕ Won't fit
72	F	Nemotron 3.5 Lightning 30B-A3B	30B	Q8_0 · 33.6 GB	—	chat · coding · reasoning	✕ Won't fit
73	F	Ornith 1.0 35B-A3B	35B	Q4_K_M · 21.2 GB	—	coding · reasoning	✕ Won't fit
74	F	XYZ-Aquila mini 35B-A3B	35B	Q4_K_M · 21.4 GB	—	chat · reasoning	✕ Won't fit
75	F	KAT-Coder V2.5 Dev	35B	Q4_K_M · 21.4 GB	—	coding · reasoning	✕ Won't fit
76	F	BigBang V1 36B-A3B	36B	Q4_K_M · 21.9 GB	—	chat · coding · vision · reasoning	✕ Won't fit
77	F	Qwen3.8 2.4T-A95B	2400B	IQ4_XS · 1310.9 GB	—	chat · coding · reasoning	✕ Won't fit
78	F	GLM-4.5V	106B	Q4_K_M · 63.6 GB	—	chat · vision · reasoning	✕ Won't fit
79	F	GLM-4.5-Air	106B	Q4_K_M · 63.6 GB	—	chat · coding · reasoning	✕ Won't fit
80	F	Apodex 1.1 Mini	35B	Q4_K_M · 21.9 GB	—	chat · coding · vision · reasoning	✕ Won't fit
81	F	Ling 3.0 Flash	124B	Q4_K_M · 77.8 GB	—	chat · coding · reasoning	✕ Won't fit
82	F	Qwen3.8-Flash-Next	125B	IQ4_XS · 93.7 GB	—	chat · coding · vision · reasoning	✕ Won't fit
83	F	GLM-5.3	744B	Q4_K_XL · 467.3 GB	—	chat · coding · reasoning	✕ Won't fit
84	F	GLM-5.3-Flash	320B	Q4_K_XL · 199.7 GB	—	chat · coding · vision · reasoning	✕ Won't fit
85	F	Llama 3.3 70B	70B	Q4_K_M · 42.5 GB	—	chat · coding	✕ Won't fit
86	F	Ornith 1.0 397B	397B	Q4_K_M · 245 GB	—	coding · reasoning	✕ Won't fit
87	F	Hunyuan 3 (Hy3)	298.8B	Q4_K_M · 182.2 GB	—	chat · coding · reasoning	✕ Won't fit
88	F	DeepSeek V4 Flash 0731	284B	Q4_K_XL · 155.1 GB	—	chat · coding · reasoning	✕ Won't fit
89	F	Inkling	952.4B	Q8_0 · 856.8 GB	—	chat · vision · reasoning	✕ Won't fit
90	F	Inkling Small	276B	Q4_K_M · 162.5 GB	—	chat · coding · vision · reasoning	✕ Won't fit
91	F	LongCat Flash Chat	561.9B	IQ1_S · 114 GB	—	chat · coding · reasoning	✕ Won't fit
92	F	Laguna XS 2.1	33B	Q4_K_M · 20.3 GB	—	coding · reasoning	✕ Won't fit
93	F	Laguna S 2.1	118B	Q4_K_M · 96 GB	—	coding · reasoning	✕ Won't fit

✓ = verified on a real device; other speeds are bandwidth-based estimates. On a phone with less RAM the ranking shifts toward smaller models — Check your own phone instead.