What's new
now data Recomputed internal benchmark aggregates now data Attempted Chatbot Arena collection now data Updated speed measurements now data Pulled latest OpenRouter price index now data Validated official pricing snapshots 1h digest Apple launches HomeView, its first smart display for the home 1h jobs Jobs market snapshot refreshed 1h digest Published the 2026-10-13 daily digest 1h digest SpaceXSI bundles X Premium, Grok and Cursor into one subscription 1h digest The final Disrupt Stage lineup: Three days of conversations you won’t hear anywhere outside of TechCrunch Disrupt 2026 now data Recomputed internal benchmark aggregates now data Attempted Chatbot Arena collection now data Updated speed measurements now data Pulled latest OpenRouter price index now data Validated official pricing snapshots 1h digest Apple launches HomeView, its first smart display for the home 1h jobs Jobs market snapshot refreshed 1h digest Published the 2026-10-13 daily digest 1h digest SpaceXSI bundles X Premium, Grok and Cursor into one subscription 1h digest The final Disrupt Stage lineup: Three days of conversations you won’t hear anywhere outside of TechCrunch Disrupt 2026

Fastest AI Models

23 LLMs ranked by speed (tokens per second). For latency-sensitive applications like chatbots, real-time coding, and interactive agents.

Note: Speed ratings are relative estimates. For production latency data, we recommend Artificial Analysis. See also: Full Speed Comparison with TTFT and provider endpoint data.

8

Ultra Fast (100+ tok/s)

12

Fast (50–99 tok/s)

2

Moderate (20–49 tok/s)

1

Slower (<20 tok/s)

Speed Ranking

1 Gemini 2.5 Flash (batch)
350 tok/s
2 GPT-4.1 Nano (batch)
200 tok/s
3 GPT-4.1 Mini (batch)
160 tok/s
4 Mistral Small 3.1 24B
150 tok/s
5 GPT-4o-mini (batch)
150 tok/s
6 Llama 4 Scout
120 tok/s
7 GPT-4.1 (batch)
110 tok/s
8 GPT-4o (batch)
100 tok/s
9 Llama 4 Maverick
95 tok/s
10 Claude Sonnet 4.6 (batch)
90 tok/s
11 Gemini 2.5 Pro
90 tok/s
12 GPT-5.2
85 tok/s
13 Claude Sonnet 4
80 tok/s
14 Mistral Large
80 tok/s
15 Llama 3.3 70B Instruct
80 tok/s
16 GPT-5 (batch)
75 tok/s
17 O4 Mini
65 tok/s
18 Qwen2.5 72B Instruct
65 tok/s
19 o3 Mini (batch)
60 tok/s
20 Claude Opus 4.6 (batch)
50 tok/s
21 DeepSeek V3.2
49 tok/s
22 DeepSeek R1
30 tok/s
23 O3
15 tok/s

Full Table

# Model Speed Price
1 Gemini 2.5 Flash (batch) Google 350 tok/s $0.15
2 GPT-4.1 Nano (batch) OpenAI 200 tok/s $0.05
3 GPT-4.1 Mini (batch) OpenAI 160 tok/s $0.20
4 Mistral Small 3.1 24B OSS Mistral 150 tok/s $0.35
5 GPT-4o-mini (batch) OpenAI 150 tok/s $0.07
6 Llama 4 Scout OSS Meta 120 tok/s $0.18
7 GPT-4.1 (batch) OpenAI 110 tok/s $1.00
8 GPT-4o (batch) OpenAI 100 tok/s $1.25
9 Llama 4 Maverick OSS Meta 95 tok/s $0.19
10 Claude Sonnet 4.6 (batch) Anthropic 90 tok/s $3.00
11 Gemini 2.5 Pro Google 90 tok/s $1.25
12 GPT-5.2 OpenAI 85 tok/s $0.88
13 Claude Sonnet 4 Anthropic 80 tok/s $3.00
14 Mistral Large OSS Mistral 80 tok/s $2.00
15 Llama 3.3 70B Instruct OSS Meta 80 tok/s $0.59
16 GPT-5 (batch) OpenAI 75 tok/s $0.63
17 O4 Mini OpenAI 65 tok/s $0.55
18 Qwen2.5 72B Instruct OSS Alibaba 65 tok/s $0.36
19 o3 Mini (batch) OpenAI 60 tok/s $0.55
20 Claude Opus 4.6 (batch) Anthropic 50 tok/s $5.00
21 DeepSeek V3.2 OSS DeepSeek 49 tok/s $0.29
22 DeepSeek R1 OSS DeepSeek 30 tok/s $0.70
23 O3 OpenAI 15 tok/s $1.00
Full leaderboard → Compare models head-to-head →