LLM Foundation Models
Compare performance, pricing, and features of leading Large Language Models
| Model & Creator | Context Window | Quality Index | Price (USD/1M) | Tokens/s | Latency (s) |
|---|---|---|---|---|---|
 o1-preview OpenAI | 128k | 86 | $27.56 | 147.5 | 22.89 |
o1-mini OpenAI | 128k | 84 | $5.25 | 225.0 | 10.21 |
 Gemini 2.0 Flash (exp) Google | 2000k | 82 | $0.00 | 168.7 | 0.53 |
 DeepSeek V3 DeepSeek | 128k | 80 | $0.48 | 89.0 | 1.00 |
Gemini 1.5 Pro (Sep) Google | 2000k | 80 | $2.19 | 60.1 | 0.82 |
 Claude 3.5 Sonnet (Oct) Anthropic | 200k | 80 | $6.00 | 67.2 | 1.02 |
GPT-4o (May '24) OpenAI | 128k | 78 | $7.50 | 102.7 | 0.63 |
GPT-4o (Aug '24) OpenAI | 128k | 78 | $4.38 | 90.5 | 0.63 |
 Qwen2.5 72B Alibaba | 131k | 77 | $0.40 | 67.0 | 0.58 |
Claude 3.5 Sonnet (June) Anthropic | 200k | 76 | $6.00 | 58.0 | 0.93 |
 Nova Pro Amazon | 300k | 75 | $1.40 | 93.1 | 0.39 |
GPT-4 Turbo OpenAI | 128k | 75 | $15.00 | 38.9 | 1.21 |
 Mistral Large 2 (Jul '24) Mistral | 128k | 74 | $3.00 | 33.4 | 0.50 |
Pixtral Large Mistral | 128k | 74 | $3.00 | 37.8 | 0.39 |
 Llama 3.1 405B Meta | 128k | 74 | $3.50 | 29.8 | 0.73 |
Llama 3.3 70B Meta | 128k | 74 | $0.67 | 72.8 | 0.48 |
GPT-4o (Nov '24) OpenAI | 128k | 73 | $4.38 | 119.9 | 0.33 |
GPT-4o mini OpenAI | 128k | 73 | $0.26 | 113.4 | 0.62 |
Gemini 1.5 Flash (Sep) Google | 1000k | 72 | $0.13 | 186.6 | 0.41 |
Claude 3 Opus Anthropic | 200k | 70 | $30.00 | 25.7 | 2.01 |
Llama 3.2 90B (Vision) Meta | 128k | 68 | $0.81 | 47.6 | 0.34 |
Llama 3.1 70B Meta | 128k | 68 | $0.72 | 72.7 | 0.46 |
Claude 3.5 Haiku Anthropic | 200k | 68 | $1.60 | 64.5 | 0.72 |
 Yi-Large 01.AI | 32k | 61 | $3.00 | 66.5 | 0.44 |
Gemma 2 27B Google | 8k | 61 | $0.26 | 48.1 | 0.75 |
Claude 3 Sonnet Anthropic | 200k | 57 | $6.00 | 66.4 | 0.74 |
 Command-R+ Cohere | 128k | 55 | $5.19 | 50.2 | 0.48 |
Gemma 2 9B Google | 8k | 55 | $0.12 | 170.0 | 0.41 |
Claude 3 Haiku Anthropic | 200k | 55 | $0.50 | 123.1 | 0.55 |
Llama 3.1 8B Meta | 128k | 54 | $0.10 | 182.9 | 0.35 |
Llama 3.2 11B (Vision) Meta | 128k | 54 | $0.18 | 132.1 | 0.30 |
Llama 3.2 3B Meta | 128k | 49 | $0.06 | 249.6 | 0.38 |
Llama 3 70B Meta | 8k | 47 | $0.89 | 49.9 | 0.40 |
Llama 3 8B Meta | 8k | 45 | $0.15 | 122.9 | 0.34 |
Llama 3.2 1B Meta | 128k | 26 | $0.04 | 468.4 | 0.38 |
GPT-4 OpenAI | 8k | - | $37.50 | 29.0 | 0.66 |
Llama 2 Chat 7B Meta | 4k | - | $0.33 | 123.9 | 0.38 |
Gemini 1.0 Pro Google | 33k | - | $0.75 | 102.8 | 1.29 |
Methodology
While higher quality models are typically more expensive, they do not all follow the same price-quality curve.
Quality Index
Average result across our evaluations covering different dimensions of model intelligence. Currently includes MMLU, GPQA, Math & HumanEval. OpenAI o1 model figures are preliminary.
Price
Price per token, represented as USD per million Tokens. Price is a blend of Input & Output token prices (3:1 ratio).
Median across providers
Figures represent median (P50) across all providers which support the model.
Data source: artificialanalysis.ai/methodology