Browse AI models

86 models tracked - grouped by family, best families first

Families Newest Quality Value Size
All Edge Consumer Enthusiast Workstation Premier
Any 8GB 12GB 24GB 32GB 48GB 64GB 96GB
More filters

Claude Fable

🇺🇸 Anthropic 2 models best 95
Claude Fable 5.1 Sep 1, 2026 text + metadata
1000k ctx no local build proprietary $10.00/M in
95
10 pts/$
Claude Fable 5 Jun 10, 2026
1000k ctx no local build proprietary $10.00/M in
95
10 pts/$

Claude Opus

🇺🇸 Anthropic 2 models best 95
Claude Opus 5 Aug 7, 2026 text + metadata
1000k ctx no local build proprietary $5.00/M in
95
19 pts/$
Claude Opus 4.8 May 29, 2026
1000k ctx no local build proprietary $5.00/M in
93
19 pts/$

GPT-5.6

🇺🇸 OpenAI 3 models best 95
GPT-5.6 Luna Aug 5, 2026 no marks
1050k ctx no local build proprietary $0.20/M in
82
410 pts/$
GPT-5.6 Sol Aug 5, 2026 no marks
1050k ctx no local build proprietary $5.00/M in
95
19 pts/$
GPT-5.6 Terra Aug 5, 2026 no marks
1050k ctx no local build proprietary $2.00/M in
90
45 pts/$

GLM / Zhipu

🇨🇳 Zhipu AI 3 models best 92
GLM-5.3-Flash MoE Aug 26, 2026
320.0B 18.0B active 1000k ctx 100GB min RAM mit 4 var $0.02/M in
91
3640 pts/$
GLM 5.3 MoE Aug 14, 2026
743.0B 40.0B active 1000k ctx no local build mit $2.80/M in
92
33 pts/$
GLM 5.2 MoE Jun 13, 2026
744.0B 40.0B active 1000k ctx 240GB min RAM mit 5 var $0.20/M in
88
440 pts/$

Grok

🇺🇸 xAI 1 model best 92
Grok 4.6 Aug 12, 2026
500k ctx no local build proprietary $2.00/M in
92
46 pts/$

Muse

🇺🇸 Meta 2 models best 92
Muse Glimmer 30B Aug 10, 2026
30.0B params 128k ctx 24GB min RAM apache 2.0 3 var
Muse Spark 1.2 Aug 5, 2026 no marks
1000k ctx no local build proprietary $1.25/M in

Claude Sonnet

🇺🇸 Anthropic 2 models best 91
Claude Sonnet 5.5 Sep 28, 2026
1000k ctx no local build proprietary $2.00/M in
91
46 pts/$
Claude Sonnet 5 Jul 1, 2026
1000k ctx no local build proprietary $2.00/M in
88
44 pts/$

Qwen

🇨🇳 Alibaba 12 models best 91
Qwen3.8-Omni-Flash MoE Sep 14, 2026
125.0B 1000k ctx no local build proprietary $0.15/M in
80
533 pts/$
Qwen3.8-Flash-Next MoE Aug 26, 2026
180.0B 6.0B active 262k ctx 24GB min RAM qwen community 1.0 3 var
89
Qwen3.8-27B Aug 5, 2026
27.0B 262k ctx 8GB min RAM apache 2.0 4 var
85
Qwen3.8-Max MoE Aug 3, 2026 text + metadata
2400.0B 95.0B active 1000k ctx no local build apache 2.0 $2.00/M in
91
46 pts/$
Qwen3.6 35B A3B MoE Apr 22, 2026 text + metadata
35.0B 3.0B active 262k ctx 20GB min RAM apache 2.0 2 var $0.05/M in
87
1740 pts/$
Qwen3.6 27B Nov 1, 2025 text + metadata
27.0B 128k ctx 17GB min RAM apache 2.0 2 var $0.30/M in
82
273 pts/$
Qwen3 14B Apr 28, 2025 text + metadata
14.7B 128k ctx 9GB min RAM apache 2.0 2 var $0.10/M in
70
700 pts/$
Qwen3 235B A22B MoE Apr 28, 2025 text + metadata
235.0B 22.0B active 128k ctx 68GB min RAM apache 2.0 2 var $0.46/M in
85
187 pts/$
Qwen3 30B A3B MoE Apr 28, 2025 text + metadata
30.5B 3.0B active 128k ctx 16GB min RAM apache 2.0 2 var $0.12/M in
72
600 pts/$
Qwen3 32B Apr 28, 2025 text + metadata
32.8B 128k ctx 20GB min RAM apache 2.0 4 var $0.08/M in
78
975 pts/$
Qwen3 4B Apr 28, 2025 text + metadata
4.0B 32k ctx 3GB min RAM apache 2.0 1 var
56
Qwen3 8B Apr 28, 2025 text + metadata
8.2B 128k ctx 5GB min RAM apache 2.0 2 var $0.12/M in
65
556 pts/$

DeepSeek

🇨🇳 DeepSeek 6 models best 90
DeepSeek V4.1 Flash MoE Sep 10, 2026
552.0B 16.0B active 1000k ctx 265GB min RAM mit 2 var $0.02/M in
90
4500 pts/$
DeepSeek V4 Flash 0731 MoE Jul 31, 2026 text + metadata
284.0B 13.0B active 1000k ctx 168GB min RAM mit 2 var $0.01/M in
90
7500 pts/$
DeepSeek V4 Pro MoE Apr 24, 2026 text + metadata
1600.0B 49.0B active 1000k ctx no local build mit $0.25/M in
89
356 pts/$
DeepSeek V3.1 Terminus MoE Sep 29, 2025 text + metadata
685.0B 37.0B active 128k ctx no local build mit $0.27/M in
82
304 pts/$
DeepSeek V3.2 Exp MoE Sep 29, 2025 text + metadata
685.0B 37.0B active 128k ctx no local build mit $0.13/M in
83
618 pts/$
DeepSeek V3 0324 MoE Mar 24, 2025 text + metadata
671.0B 37.0B active 128k ctx 188GB min RAM mit 2 var $0.24/M in
87
363 pts/$

Other labs (Kimi · LFM · Command R+)

5 models best 90
Kimi K3 MoE Moonshot AI Jul 16, 2026 text + metadata
2800.0B 104.0B active 1000k ctx 610GB min RAM other 1 var $0.68/M in
90
132 pts/$
Kimi K2.7 Code MoE Moonshot AI Jun 16, 2026 text + metadata
1000.0B 32.0B active 256k ctx no local build other $0.66/M in
86
131 pts/$
LFM2.5 8B A1B MoE Liquid AI Sep 1, 2025
8.0B 1.0B active 32k ctx 3GB min RAM apache 2.0 1 var
58
Kimi K2 MoE Moonshot AI Jul 11, 2025 text + metadata
1000.0B 128k ctx no local build other $0.57/M in
85
149 pts/$
Command R+ 104B Cohere Apr 4, 2024
104.0B 128k ctx no local build cc by 4.0 $2.50/M in
75
30 pts/$

Qwen Image

🇨🇳 Alibaba 1 model best 90
Qwen-Image-2.1 Sep 20, 2026
7.0B 16GB min RAM qwen research 1 var

Ornith

🇺🇸 Ornith AI 3 models best 89
Ornith-1.5-35B-A3B MoE Aug 18, 2026
36.0B 3.0B active 262k ctx 22GB min RAM mit 5 var
86
Ornith-1.5-397B MoE Aug 18, 2026
403.0B 262k ctx 248GB min RAM mit 4 var
89
Ornith-1.5-9B Aug 18, 2026
9.0B 262k ctx 6GB min RAM mit 5 var
74

Pokee

🇺🇸 Pokee AI 1 model best 88
Pokee-Isaac 28B Aug 4, 2026
28.0B params 10000k ctx 18GB min RAM other 3 var $0.15/M in
88
587 pts/$

Gemini

🇺🇸 Google 1 model best 87
Gemini 3.7 Flash Aug 13, 2026
1000k ctx no local build proprietary $0.75/M in
87
116 pts/$

LongCat

🇨🇳 Meituan 1 model best 86
LongCat 2.0 MoE Jul 3, 2026
1600.0B 48.0B active 1000k ctx no local build mit $0.30/M in
86
287 pts/$

Ember

🇺🇸 Fireworks Research 1 model best 85
Ember-1 Sep 23, 2026
no local build proprietary
85

Gemma

🇺🇸 Google 10 models best 85
Gemma 4 26B A4B MoE Apr 30, 2026 no marks
25.2B 3.8B active 256k ctx 15GB min RAM other 4 var $0.04/M in
84
2000 pts/$
Gemma 4 31B Apr 2, 2026 no marks
30.7B 256k ctx 19GB min RAM apache 2.0 5 var $0.08/M in
85
1063 pts/$
Gemma 3 270M Aug 14, 2025 no marks
270M 32k ctx 1GB min RAM other 2 var
25
Gemma 3n E2B Jun 26, 2025 no marks
2.0B 32k ctx 6GB min RAM other 2 var
48
Gemma 3n E4B Jun 26, 2025 no marks
4.0B 32k ctx 8GB min RAM other 2 var
58
Gemma 4 12B Jun 1, 2025 no marks
12.0B 128k ctx 7GB min RAM other 2 var
72
Gemma 3 12B Feb 27, 2025 no marks
12.2B 128k ctx 8GB min RAM other 2 var $0.05/M in
68
1360 pts/$
Gemma 3 1B Feb 27, 2025 no marks
1.0B 32k ctx 1GB min RAM other 2 var
38
Gemma 3 27B Feb 27, 2025 no marks
27.2B 128k ctx 17GB min RAM other 2 var $0.08/M in
72
900 pts/$
Gemma 3 4B Feb 27, 2025 no marks
4.3B 128k ctx 3GB min RAM other 2 var $0.05/M in
55
1100 pts/$

Nemotron

🇺🇸 NVIDIA 4 models best 85
Nemotron 3 Diarization Sep 23, 2026
100M params 0GB min RAM openmdw 1.1 2 var
Nemotron 3.5 Lightning MoE Aug 11, 2026
31.6B 3.6B active 1000k ctx 34GB min RAM other 3 var
74
Nemotron 3 Nano 4B Jun 5, 2026
4.0B 1000k ctx 3GB min RAM other 1 var
Nemotron 3 Ultra MoE Jun 4, 2026
550.0B 55.0B active 262k ctx no local build other $0.50/M in
85
170 pts/$

Swe

🇺🇸 Cognition 1 model best 85
SWE-2 MoE Sep 10, 2026
2800.0B 104.0B active no local build proprietary
85

Union Alpha (stealth, operator unconfirmed)

1 model best 85
Union Alpha MoE Sep 16, 2026
no local build proprietary
85

Llama

🇺🇸 Meta 5 models best 84
Llama 3.3 70B Instruct Dec 6, 2024 no marks
70.6B 128k ctx 22GB min RAM llama 3 4 var $0.10/M in
80
800 pts/$
Llama 3.2 1B Instruct Sep 25, 2024 no marks
1.2B 128k ctx 1GB min RAM llama 3 2 var $0.03/M in
35
1296 pts/$
Llama 3.2 3B Instruct Sep 25, 2024 no marks
3.2B 128k ctx 2GB min RAM llama 3 2 var $0.05/M in
50
1000 pts/$
Llama 3.1 405B Jul 23, 2024 no marks
405.0B 128k ctx no local build llama 3
84
Llama 3.1 8B Instruct Jul 23, 2024 no marks
8.0B 128k ctx 5GB min RAM llama 3 3 var $0.02/M in
62
3100 pts/$

Mimo

🇨🇳 Xiaomi 2 models best 84
MiMo-V2.6-Flash MoE Sep 21, 2026
309.0B 15.0B active 1024k ctx 80GB min RAM mit 2 var
76
MiMo-V2.6-Pro MoE Sep 21, 2026
1020.0B 42.0B active 1024k ctx 2040GB min RAM mit 1 var
84

Jeff

mstrasser (firelex) 1 model best 82
Jeff Sep 28, 2026
2.0B params 262k ctx 1GB min RAM apache 2.0 2 var

Flux

🇩🇪 Black Forest Labs 1 model best 78
Flux 3 Action Sep 24, 2026
7.0B params 17GB min RAM flux kommunity 1 var

Xing

🇨🇳 China Telecom AI 1 model best 72
Xing4.0-29B-A4B MoE Sep 22, 2026
29.0B 4.0B active 256k ctx 15GB min RAM apache 2.0 2 var
72

Phi

🇺🇸 Microsoft 2 models best 70
Phi-4 Mini Feb 1, 2025 no marks
3.8B 128k ctx 3GB min RAM mit 2 var
52
Phi-4 14B Dec 12, 2024 no marks
14.7B 16k ctx 9GB min RAM mit 3 var $0.07/M in
70
1000 pts/$

Mistral

🇫🇷 Mistral AI 2 models best 68
Mistral Nemo 12B Jul 18, 2024
12.2B 128k ctx 7GB min RAM apache 2.0 2 var $0.02/M in
62
3444 pts/$
Mixtral 8x7B Instruct MoE Dec 11, 2023
46.7B 12.9B active 32k ctx 19GB min RAM apache 2.0 2 var
68

Gliner

🇺🇸 Fastino 1 model best 64
GLiNER2.5-Decide Sep 24, 2026
340M params 1GB min RAM apache 2.0 2 var

Pareto

🇺🇸 Unbiased AI 1 model best 40
Pareto Sep 17, 2026
no local build proprietary hosted

Cua S1

trycua 1 model best 0
CUA-S1-FORMS Sep 18, 2026
no local build mit (source only no weights published)

Jev

TypeSafe 1 model best 0
Jev Sep 15, 2026 no marks
60k ctx no local build proprietary $0.04/M in

K2 Horizon

IFM Technologies 1 model best 0
K2 Horizon 375B Sep 3, 2026
375.0B params no local build

Laya

ConvAI Innovations 1 model best 0
Laya Sep 18, 2026 no marks
420M params 1k ctx 1GB min RAM apache 2.0 3 var

Needle

Cactus Compute 1 model best 0
Needle 3 Sep 17, 2026
120M params 8k ctx no local build

Nex N2 5

Nex AGI 1 model best 0
Nex-N2.5-Max MoE Sep 8, 2026
1600.0B 49.0B active 1050k ctx no local build

Openjev

Undisclosed 1 model best 0
OpenJev Sep 16, 2026 no marks
600M params 8k ctx 1GB min RAM mit 3 var

Ternary Bonsai

Prism ML 1 model best 0
Ternary Bonsai 2 27B Sep 17, 2026 no marks
27.4B params 262k ctx no local build apache 2.0

FAQ

What is the cheapest LLM API?

Open-weight models like DeepSeek, Qwen, and GLM are the cheapest LLM APIs - often under $1 per million input tokens. Sort the table by best value to rank quality per dollar.

Which LLM has the best coding score?

Coding scores vary by benchmark; filter the table by use case coding to see coding_score per model, or open any model's page for its benchmark breakdown.

How do I compare LLM pricing?

This table shows per-million-token input and output prices across providers. Open a model's pricing page for live rates, cached-input discounts, and price history.