LLM Leaderboard 2026

321+ AI models ranked by our RunFree Score — a 0-100 composite of reasoning, coding, math and agentic benchmarks — alongside live API pricing and context windows. Right now, Qwen3 Max Thinking leads at 94.8.

Pricing updated June 28, 2026 · 13 models scored

Most intelligent
Qwen3 Max Thinking
RunFree 94.8
Best value
DeepSeek V4 Flash
59.1 · $0.11/1M
Cheapest paid
Ling-2.6-flash
$0.01/1M tokens
Largest context
Llama 4 Scout
10M tokens

Full ranked, filterable table of AI models

#Model
1Qwen3 Max Thinking
94.8
$1.56
2Claude Opus 4.8
94.7
$10.00
3GLM 5.2
86.3
$1.46
4Claude Opus 4.7
85.7
$10.00
5GPT-5.5
78.8
$11.25
6Gemini 3.1 Pro Preview
78
$4.50
7DeepSeek V4 Pro
72.4
$0.54
8Claude Sonnet 4.6
66.2
$6.00
9Gemini 3 Flash Preview
63.5
$1.13
10DeepSeek V4 Flash
59.1
$0.11
11Kimi K2 Thinking
45.2
$1.07
12Gemini 2.5 Pro
22.3
$3.44
13Llama 4 Maverick
0
$0.26
14Aion-1.0
$5.00
15Aion-1.0-Mini
$0.87
16Aion-2.0
$1.00
17Aion-RP 1.0 (8B)
$1.00
18Anthropic Claude Haiku Latest
$2.00
19Anthropic Claude Sonnet Latest
$6.00
20Claude 3 Haiku
$0.50
21Claude Fable 5
$20.00
22Claude Fable Latest
$20.00
23Claude Haiku 4.5
$2.00
24Claude Opus 4
$30.00
25Claude Opus 4.1
$30.00
26Claude Opus 4.5
$10.00
27Claude Opus 4.6
$10.00
28Claude Opus 4.6 (Fast)
$60.00
29Claude Opus 4.7 (Fast)
$60.00
30Claude Opus 4.8 (Fast)
$20.00
31Claude Opus Latest
$10.00
32Claude Sonnet 4
$6.00
33Claude Sonnet 4.5
$6.00
34Coder Large
$0.57
35Codestral 2508
$0.45
36Cogito v2.1 671B
$1.25
37Command A
$4.38
38Command R (08-2024)
$0.26
39Command R+ (08-2024)
$4.38
40Command R7B (12-2024)
$0.07
41Cydonia 24B V4.1
$0.35
42DeepSeek V3
$0.35
43DeepSeek V3 0324
$0.34
44DeepSeek V3.1
$0.35
45DeepSeek V3.1 Terminus
$0.44
46DeepSeek V3.2
$0.26
47DeepSeek V3.2 Exp
$0.30
48Devstral 2 2512
$0.80
49ERNIE 4.5 VL 424B A47B
$0.63
50Fugu Ultra
$11.25
51Gemini 2.5 Flash
$0.85
52Gemini 2.5 Flash Lite
$0.17
53Gemini 2.5 Flash Lite Preview 09-2025
$0.17
54Gemini 2.5 Pro Preview 05-06
$3.44
55Gemini 2.5 Pro Preview 06-05
$3.44
56Gemini 3.1 Flash Lite
$0.56
57Gemini 3.1 Flash Lite Preview
$0.56
58Gemini 3.1 Pro Preview Custom Tools
$4.50
59Gemini 3.5 Flash
$3.38
60Gemma 2 27B
$0.65
61Gemma 3 12B
$0.07
62Gemma 3 27B
$0.10
63Gemma 3 4B
$0.06
64Gemma 3n 4B
$0.07
65Gemma 4 26B A4B
$0.13
66Gemma 4 26B A4B (free)
Free
67Gemma 4 31B
$0.18
68Gemma 4 31B (free)
Free
69GLM 4.5
$1.00
70GLM 4.5 Air
$0.31
71GLM 4.5V
$0.90
72GLM 4.6
$0.76
73GLM 4.6V
$0.45
74GLM 4.7
$0.74
75GLM 4.7 Flash
$0.14
76GLM 5
$0.93
77GLM 5 Turbo
$1.90
78GLM 5.1
$1.50
79GLM 5V Turbo
$1.90
80Google Gemini Flash Latest
$3.38
81Google Gemini Pro Latest
$4.50
82GPT Chat Latest
$11.25
83GPT-3.5 Turbo
$0.75
84GPT-3.5 Turbo (older v0613)
$1.25
85GPT-3.5 Turbo 16k
$3.25
86GPT-3.5 Turbo Instruct
$1.63
87GPT-4
$37.50
88GPT-4 Turbo
$15.00
89GPT-4 Turbo Preview
$15.00
90GPT-4.1
$3.50
91GPT-4.1 Mini
$0.70
92GPT-4.1 Nano
$0.17
93GPT-4o
$4.38
94GPT-4o (2024-05-13)
$7.50
95GPT-4o (2024-08-06)
$4.38
96GPT-4o (2024-11-20)
$4.38
97GPT-4o Search Preview
$4.38
98GPT-4o-mini
$0.26
99GPT-4o-mini (2024-07-18)
$0.26
100GPT-4o-mini Search Preview
$0.26
101GPT-5
$3.44
102GPT-5 Chat
$3.44
103GPT-5 Codex
$3.44
104GPT-5 Mini
$0.69
105GPT-5 Nano
$0.14
106GPT-5 Pro
$41.25
107GPT-5.1
$3.44
108GPT-5.1 Chat
$3.44
109GPT-5.1-Codex
$3.44
110GPT-5.1-Codex-Max
$3.44
111GPT-5.1-Codex-Mini
$0.69
112GPT-5.2
$4.81
113GPT-5.2 Chat
$4.81
114GPT-5.2 Pro
$57.75
115GPT-5.2-Codex
$4.81
116GPT-5.3 Chat
$4.81
117GPT-5.3-Codex
$4.81
118GPT-5.4
$5.63
119GPT-5.4 Mini
$1.69
120GPT-5.4 Nano
$0.46
121GPT-5.4 Pro
$67.50
122GPT-5.5 Pro
$67.50
123gpt-oss-120b
$0.06
124gpt-oss-120b (free)
Free
125gpt-oss-20b
$0.06
126gpt-oss-20b (free)
Free
127gpt-oss-safeguard-20b
$0.13
128Granite 4.0 Micro
$0.04
129Granite 4.1 8B
$0.06
130Grok 4.20
$1.56
131Grok 4.20 Multi-Agent
$1.56
132Grok 4.3
$1.56
133Grok Build 0.1
$1.25
134Hermes 3 405B Instruct
$1.00
135Hermes 3 405B Instruct (free)
Free
136Hermes 3 70B Instruct
$0.70
137Hermes 4 405B
$1.50
138Hermes 4 70B
$0.20
139Hunyuan A13B Instruct
$0.25
140Hy3 preview
$0.10
141Inflection 3 Pi
$4.38
142Inflection 3 Productivity
$4.38
143Jamba Large 1.7
$3.50
144KAT-Coder-Pro V2
$0.52
145Kimi K2 0711
$1.00
146Kimi K2 0905
$1.07
147Kimi K2.5
$0.79
148Kimi K2.6
$1.35
149Kimi K2.7 Code
$1.43
150Laguna M.1
$0.25
151Laguna M.1 (free)
Free
152Laguna XS.2
$0.13
153Laguna XS.2 (free)
Free
154LFM2-24B-A2B
$0.05
155LFM2.5-1.2B-Instruct (free)
Free
156LFM2.5-1.2B-Thinking (free)
Free
157Ling-2.6-1T
$0.21
158Ling-2.6-flash
$0.01
159Llama 3 8B Instruct
$0.14
160Llama 3 8B Lunaris
$0.04
161Llama 3.1 70B Hanami x1
$3.00
162Llama 3.1 70B Instruct
$0.40
163Llama 3.1 8B Instruct
$0.02
164Llama 3.1 Euryale 70B v2.2
$0.85
165Llama 3.2 11B Vision Instruct
$0.34
166Llama 3.2 1B Instruct
$0.07
167Llama 3.2 3B Instruct
$0.12
168Llama 3.2 3B Instruct (free)
Free
169Llama 3.3 70B Instruct
$0.15
170Llama 3.3 70B Instruct (free)
Free
171Llama 3.3 Euryale 70B
$0.68
172Llama 3.3 Nemotron Super 49B V1.5
$0.40
173Llama 4 Scout
$0.15
174Llama Guard 4 12B
$0.18
175Magnum v4 72B
$3.50
176Mercury 2
$0.38
177MiMo-V2.5
$0.15
178MiMo-V2.5-Pro
$0.54
179MiniMax M1
$0.85
180MiniMax M2
$0.44
181MiniMax M2-her
$0.52
182MiniMax M2.1
$0.45
183MiniMax M2.5
$0.21
184MiniMax M2.7
$0.32
185MiniMax M3
$0.52
186MiniMax-01
$0.43
187Ministral 3 14B 2512
$0.20
188Ministral 3 3B 2512
$0.10
189Ministral 3 8B 2512
$0.15
190Mistral Large
$3.00
191Mistral Large 2407
$3.00
192Mistral Large 3 2512
$0.75
193Mistral Medium 3
$0.80
194Mistral Medium 3.1
$0.80
195Mistral Medium 3.5
$3.00
196Mistral Nemo
$0.02
197Mistral Small 3
$0.06
198Mistral Small 3.1 24B
$0.40
199Mistral Small 3.2 24B
$0.11
200Mistral Small 4
$0.26
201Mixtral 8x22B Instruct
$3.00
202MoonshotAI Kimi Latest
$1.35
203Morph V3 Fast
$0.90
204Morph V3 Large
$1.15
205MythoMax 13B
$0.06
206Nemotron 3 Nano 30B A3B
$0.09
207Nemotron 3 Nano 30B A3B (free)
Free
208Nemotron 3 Nano Omni (free)
Free
209Nemotron 3 Super
$0.16
210Nemotron 3 Super (free)
Free
211Nemotron 3 Ultra
$0.93
212Nemotron 3 Ultra (free)
Free
213Nemotron 3.5 Content Safety (free)
Free
214Nemotron Nano 12B 2 VL (free)
Free
215Nemotron Nano 9B V2 (free)
Free
216Nex-N2-Pro
$0.44
217North Mini Code (free)
Free
218Nova 2 Lite
$0.85
219Nova Lite 1.0
$0.10
220Nova Micro 1.0
$0.06
221Nova Premier 1.0
$5.00
222Nova Pro 1.0
$1.40
223o1
$26.25
224o1-pro
$263
225o3
$3.50
226o3 Deep Research
$17.50
227o3 Mini
$1.93
228o3 Mini High
$1.93
229o3 Pro
$35.00
230o4 Mini
$1.93
231o4 Mini Deep Research
$3.50
232o4 Mini High
$1.93
233Olmo 3 32B Think
$0.24
234OpenAI GPT Latest
$11.25
235OpenAI GPT Mini Latest
$1.69
236Palmyra X5
$1.95
237Perceptron Mk1
$0.49
238Phi 4
$0.09
239Phi 4 Mini Instruct
$0.15
240Qwen Plus 0728
$0.39
241Qwen Plus 0728 (thinking)
$0.39
242Qwen-Plus
$0.39
243Qwen2.5 72B Instruct
$0.37
244Qwen2.5 7B Instruct
$0.05
245Qwen2.5 Coder 32B Instruct
$0.74
246Qwen2.5 VL 72B Instruct
$0.85
247Qwen3 14B
$0.14
248Qwen3 235B A22B
$0.80
249Qwen3 235B A22B Instruct 2507
$0.09
250Qwen3 235B A22B Thinking 2507
$0.10
251Qwen3 30B A3B
$0.21
252Qwen3 30B A3B Instruct 2507
$0.08
253Qwen3 30B A3B Thinking 2507
$0.16
254Qwen3 32B
$0.13
255Qwen3 8B
$0.14
256Qwen3 Coder 30B A3B Instruct
$0.12
257Qwen3 Coder 480B A35B
$0.61
258Qwen3 Coder 480B A35B (free)
Free
259Qwen3 Coder Flash
$0.39
260Qwen3 Coder Next
$0.28
261Qwen3 Coder Plus
$1.30
262Qwen3 Max
$1.56
263Qwen3 Next 80B A3B Instruct
$0.34
264Qwen3 Next 80B A3B Instruct (free)
Free
265Qwen3 Next 80B A3B Thinking
$0.27
266Qwen3 VL 235B A22B Instruct
$0.37
267Qwen3 VL 235B A22B Thinking
$0.84
268Qwen3 VL 30B A3B Instruct
$0.23
269Qwen3 VL 30B A3B Thinking
$0.49
270Qwen3 VL 32B Instruct
$0.18
271Qwen3 VL 8B Instruct
$0.18
272Qwen3 VL 8B Thinking
$0.43
273Qwen3.5 397B A17B
$0.90
274Qwen3.5 Plus 2026-02-15
$0.58
275Qwen3.5 Plus 2026-04-20
$0.67
276Qwen3.5-122B-A10B
$0.72
277Qwen3.5-27B
$0.54
278Qwen3.5-35B-A3B
$0.35
279Qwen3.5-9B
$0.11
280Qwen3.5-Flash
$0.11
281Qwen3.6 27B
$0.79
282Qwen3.6 35B A3B
$0.35
283Qwen3.6 Flash
$0.42
284Qwen3.6 Max Preview
$2.34
285Qwen3.6 Plus
$0.73
286Qwen3.7 Max
$1.88
287Qwen3.7 Plus
$0.56
288R1
$1.15
289R1 0528
$0.91
290R1 Distill Llama 70B
$0.80
291Reka Edge
$0.10
292Reka Flash 3
$0.13
293Relace Apply 3
$0.95
294Relace Search
$1.50
295ReMM SLERP 13B
$0.50
296Ring-2.6-1T
$0.21
297Rocinante 12B
$0.31
298Saba
$0.30
299Seed 1.6
$0.69
300Seed 1.6 Flash
$0.13
301Seed-2.0-Lite
$0.69
302Seed-2.0-Mini
$0.17
303Skyfall 36B V2
$0.61
304Solar Pro 3
$0.26
305Sonar
$1.00
306Sonar Deep Research
$3.50
307Sonar Pro
$6.00
308Sonar Pro Search
$6.00
309Sonar Reasoning Pro
$3.50
310Step 3.5 Flash
$0.14
311Step 3.7 Flash
$0.44
312Switchpoint Router
$1.49
313Trinity Large Thinking
$0.39
314Trinity Mini
$0.07
315UI-TARS 7B
$0.13
316Uncensored (free)
Free
317UnslopNemo 12B
$0.40
318Virtuoso Large
$0.86
319Voxtral Small 24B 2507
$0.15
320Weaver (alpha)
$0.81
321WizardLM-2 8x22B
$0.62

Showing 321 of 321 models. Click a column to sort, a model to open its card.

Models without enough public benchmark data are unscored. Speed (tok/s) shows where measured. Data as of June 28, 2026; see methodology.

Top 15 by RunFree Score

Loading chart…

Intelligence vs price

Loading chart…

Up and to the left is better: smarter for less. Log price scale.

Explore by category

AI model leaderboard — FAQ

Which is the most intelligent AI model right now?

By our RunFree Score, Qwen3 Max Thinking ranks highest at 94.8/100, a composite of reasoning, coding, math and agentic benchmarks. See the full ranking above.

What is the best value AI model (smart and cheap)?

DeepSeek V4 Flash offers one of the best intelligence-per-dollar ratios, scoring 59.1 at about $0.11/1M tokens (blended). The intelligence-vs-price chart above shows the full picture.

Which is the cheapest AI model?

Ling-2.6-flash is among the cheapest paid models at roughly $0.01/1M tokens (blended 3:1 input:output). Several models are also free to use.

Which model has the largest context window?

Llama 4 Scout has one of the largest context windows in the list. Sort the table by Context to compare them all.

How are these AI models scored?

Each model gets a RunFree Score: we normalize its results on public benchmarks (GPQA, SWE-Bench, AIME and more), average them within reasoning, coding, math and agentic categories, then weight the categories. The full formula is on our methodology page.

How often is this leaderboard updated?

Model details and API pricing refresh automatically every hour from OpenRouter. Benchmark scores are added from each model's official source as it publishes them, with the date shown.