709 models

Compare AI Models by Intelligence, Speed and Cost

Compare intelligence scores, output speed (tokens/s), first-chunk latency and pricing. Data from Artificial Analysis.

#1 Intelligence Score57.6
Claude Opus 5.5 (Max, Default Fallback)Anthropic · Strong results on complex, multi-step tasks
Fastest Generation1,697.3 tokens/s
Celeris-1Celeris · High-throughput streaming output
Lowest Price$0.04 / 1M tokens
Gemma 4 E4B (Reasoning)Google
#1 Open-Weight Model40.2
Qwen3.8 MaxAlibaba · A leading score among open-weight models
RankModel and provider
Intelligence score
Output speed (tokens/s)
Price per 1M tokens (3:1 blended)
Latency (first chunk)
Context length
1
Claude Opus 5.5 (Max, Default Fallback)
Anthropic
57.6
95.1tokens/s
$8.00/ 1M tokens
405.54s
--
2
Claude Sonnet 5.5 (Max, Default Fallback)
Anthropic
56
136.4tokens/s
$4.00/ 1M tokens
261.38s
--
3
Claude Opus 5.5 (Xhigh, Default Fallback)
Anthropic
56
73.7tokens/s
$8.00/ 1M tokens
48.99s
--
4
Muse Wonder Full (Max)
Meta
55.8
--------
5
Kimi K3.1 Pro
Kimi
55.1
56.5tokens/s
$4.00/ 1M tokens
1.87s
--
6
Claude Opus 5.5 (High, Default Fallback)
Anthropic
53.6
71.9tokens/s
$8.00/ 1M tokens
14.17s
--
7
Claude Fable 5.1 (Max, Default Fallback)
Anthropic
53.4
69.9tokens/s
$20.00/ 1M tokens
125.09s
--
8
Claude Fable 5.1 (Xhigh, Default Fallback)
Anthropic
53.2
52.7tokens/s
$20.00/ 1M tokens
30.98s
--
9
GPT-6 Astra (Max)
OpenAI
52.7
60.1tokens/s
$20.00/ 1M tokens
166.2s
--
10
Gemini 4 Argon (High)
Google
52.6
79.3tokens/s
$4.00/ 1M tokens
30.45s
--
11
Muse Wonder (Max)
Meta
52.4
135.3tokens/s
$8.00/ 1M tokens
30.86s
--
12
GPT-6 Astra (Xhigh)
OpenAI
52.4
54.8tokens/s
$20.00/ 1M tokens
61.84s
--
13
Claude Sonnet 5.5 (Xhigh, Default Fallback)
Anthropic
51.9
95.4tokens/s
$4.00/ 1M tokens
11.45s
--
14
GPT-6.1 Sol (Max)
OpenAI
51.8
55.8tokens/s
$4.00/ 1M tokens
132.36s
--
15
Claude Opus 5.5 (Medium, Default Fallback)
Anthropic
51.2
69tokens/s
$8.00/ 1M tokens
10.11s
--
16
Claude Fable 5.1 (High, Default Fallback)
Anthropic
51.2
49.5tokens/s
$20.00/ 1M tokens
2.4s
--
17
GPT-6.1 Sol (Xhigh)
OpenAI
51
52.2tokens/s
$4.00/ 1M tokens
53.48s
--
18
GPT-6 Astra (High)
OpenAI
50.9
49.9tokens/s
$20.00/ 1M tokens
21.37s
--
19
Claude Opus 5 (Max)Reasoning
Anthropic
50.8
--
$10.00/ 1M tokens
--1M
20
GPT-6.1 Sol (High)
OpenAI
50.2
47.3tokens/s
$4.00/ 1M tokens
20.67s
--
21
Claude Opus 5 (Xhigh)Reasoning
Anthropic
49.7
--
$10.00/ 1M tokens
--1M
22
Claude Fable 5 (Max, Opus 4.8 Fallback)
Anthropic
49.6
--
$20.00/ 1M tokens
----
23
GPT-6 Astra (Medium)
OpenAI
49.6
49.5tokens/s
$20.00/ 1M tokens
2.61s
--
24
Claude Fable 5.1 (Medium, Default Fallback)
Anthropic
48.9
48.7tokens/s
$20.00/ 1M tokens
2.05s
--
25
Claude Opus 5 (High)Reasoning
Anthropic
48.1
--
$10.00/ 1M tokens
--1M
26
Muse Spark 1.3 (Max)
Meta
48.1
--
$2.00/ 1M tokens
----
27
GPT-6.1 Sol (Medium)
OpenAI
47.8
48.6tokens/s
$4.00/ 1M tokens
2.76s
--
28
GPT-6 Sol (Max)
OpenAI
47.6
--
$4.00/ 1M tokens
----
29
GPT-5.6 Sol (Max)Reasoning
OpenAI
47
--
$8.00/ 1M tokens
--1M
30
Claude Fable 5.1 (Low, Default Fallback)
Anthropic
46.8
51.3tokens/s
$20.00/ 1M tokens
1.02s
--
31
Claude Sonnet 5.5 (High, Default Fallback)
Anthropic
46.8
92.5tokens/s
$4.00/ 1M tokens
7.75s
--
32
Grok 4.7 (Xhigh)
SpaceXAI
46.4
86tokens/s
$3.00/ 1M tokens
15.46s
--
33
Grok 4.7 (High)
SpaceXAI
46.3
79.7tokens/s
$3.00/ 1M tokens
9.87s
--
34
MiMo-V2.6-Pro
Xiaomi
46.3
44.7tokens/s
$0.54/ 1M tokens
2.36s
--
35
GPT-6 Astra (Low)
OpenAI
45.8
54.1tokens/s
$20.00/ 1M tokens
1.59s
--
36
Qwen3.8 Max (0902)
Alibaba
45.4
40.2tokens/s
$3.00/ 1M tokens
1.92s
--
37
DeepSeek V4.1 Flash 1017 (Max)
DeepSeek
45.1
173.7tokens/s
$0.53/ 1M tokens
0.95s
--
38
Muse Spark 1.3 (Xhigh)
Meta
45.1
--
$2.00/ 1M tokens
----
39
Claude Opus 5 (Medium)Reasoning
Anthropic
44.8
--
$10.00/ 1M tokens
--1M
40
GLM-5.3 (Max)Reasoning
Z AI
44.8
76.3tokens/s
$2.15/ 1M tokens
2.51s
1M

Data source and notes

Data source: Artificial Analysis · Updated Oct 25, 2026

Intelligence scores, speed, latency and prices are public measurements from Artificial Analysis. Dora Meow organizes them for comparison. First-chunk latency is the time from sending a request to receiving the first response chunk; for reasoning models it may include thinking time. Context length, reasoning and open-weight labels appear only when they match existing site records.