Flagship AI Models — Pricing & Comparison
The most powerful AI models from leading providers. Best for complex creative work, nuanced analysis, and tasks where quality is paramount.
All Flagship Models — Sorted by Price
| # | Model | Provider | Input $/M | Output $/M | Context | 1M Requests* |
|---|---|---|---|---|---|---|
| 1 | Llama 4 Maverick 2025-04-05 | Meta (via Together AI) | $0.27 | $0.85 | 1000K | $$0.000695 |
| 2 | Mistral Large 3 2025-12-02 | Mistral AI | $0.5 | $1.5 | 256K | $$0.001250 |
| 3 | DeepSeek V4 Pro 2026-04-24 | DeepSeek | $0.66 | $1.98 | 1000K | $$0.001650 |
| 4 | Gemini 3.8 Flash 2026-09-02 | $0.75 | $3.75 | 1049K | $$0.002625 | |
| 5 | Gemini 3.7 Flash 2026-08-13 | $0.75 | $3.75 | 1000K | $$0.002625 | |
| 6 | GPT-5.1 2025-11-12 | OpenAI | $1.25 | $10 | 1000K | $$0.006250 |
| 7 | GPT-5 2025-08-07 | OpenAI | $1.25 | $10 | 1000K | $$0.006250 |
| 8 | Gemini 2.5 Pro 2025-03-25 | $1.25 | $10 | 2000K | $$0.006250 | |
| 9 | Gemini Omni 1.1 Flash 2026-08-27 | $1.5 | $17.5 | 1049K | $$0.01025 | |
| 10 | Gemini 3.5 Flash 2026-05-19 | $1.5 | $9 | 1000K | $$0.006000 | |
| 11 | Mistral Medium 3.5 2026-04-29 | Mistral AI | $1.5 | $7.5 | 256K | $$0.005250 |
| 12 | GPT-5.2 2025-12-11 | OpenAI | $1.75 | $14 | 1000K | $$0.008750 |
| 13 | GPT-4.1 2025-04-14 | OpenAI | $2 | $8 | 200K | $$0.006000 |
| 14 | Gemini 3.1 Pro 2026-02-19 | $2 | $12 | 1000K | $$0.008000 | |
| 15 | Gemini 3 Pro 2025-11-18 | $2 | $12 | 2000K | $$0.008000 | |
| 16 | Grok 4.7 2026-09-21 | xAI | $2 | $6 | 500K | $$0.005000 |
| 17 | Grok 4.6 2026-08-12 | xAI | $2 | $6 | 500K | $$0.005000 |
| 18 | Grok 4.5 2026-07-08 | xAI | $2 | $6 | 500K | $$0.005000 |
| 19 | GPT-5.4 2026-03-06 | OpenAI | $2.5 | $15 | 1050K | $$0.01 |
| 20 | GPT-4o 2024-05-13 | OpenAI | $2.5 | $10 | 128K | $$0.007500 |
| 21 | Command R+ 2024-04-04 | Cohere | $2.5 | $10 | 128K | $$0.007500 |
| 22 | Grok 3 2025-02-17 | xAI | $3 | $15 | 131K | $$0.0105 |
| 23 | Llama 3.1 405B 2024-07-23 | Meta (via Together AI) | $3.5 | $3.5 | 128K | $$0.005250 |
| 24 | GPT-5.6 Sol 2026-06-26 | OpenAI | $5 | $30 | 1050K | $$0.02 |
| 25 | GPT-5.5 2026-04-24 | OpenAI | $5 | $30 | 1050K | $$0.02 |
| 26 | GPT-5.5 Instant 2026-05-05 | OpenAI | $5 | $30 | 1050K | $$0.02 |
| 27 | Claude Opus 5 2026-07-24 | Anthropic | $5 | $25 | 1000K | $$0.0175 |
| 28 | Claude Opus 4.7 2026-04-16 | Anthropic | $5 | $25 | 1000K | $$0.0175 |
| 29 | Claude Opus 4.6 2026-02-05 | Anthropic | $5 | $25 | 1000K | $$0.0175 |
| 30 | Claude Opus 4.5 2025-11-01 | Anthropic | $5 | $25 | 200K | $$0.0175 |
| 31 | GPT-6 Astra 2026-09-03 | OpenAI | $10 | $50 | 1050K | $$0.035 |
| 32 | GPT-4 Turbo 2024-04-09 | OpenAI | $10 | $30 | 128K | $$0.025 |
| 33 | Claude Fable 5.1 2026-09-01 | Anthropic | $10 | $50 | 1000K | $$0.035 |
| 34 | Claude Mythos 5.1 2026-09-01 | Anthropic | $10 | $50 | 1000K | $$0.035 |
| 35 | Claude Fable 5 2026-06-09 | Anthropic | $10 | $50 | 1000K | $$0.035 |
| 36 | GPT-5 Pro 2025-08-07 | OpenAI | $15 | $120 | 200K | $$0.075 |
| 37 | Claude Opus 4 2025-05-14 | Anthropic | $15 | $75 | 200K | $$0.0525 |
| 38 | Claude Opus 4.1 2025-08-05 | Anthropic | $15 | $75 | 200K | $$0.0525 |
| 39 | Claude 3 Opus 2024-03-04 | Anthropic | $15 | $75 | 200K | $$0.0525 |
| 40 | o1 Pro 2024-12-17 | OpenAI | $150 | $600 | 200K | $$0.45 |
* Estimated cost for 1M requests at 1,000 input + 500 output tokens each.
Calculate Your Flagship Model Costs
💡 Tips for Using Flagship Models
Use flagship models for your most important tasks — creative writing, complex analysis, customer-facing content
Consider using a cheaper model for preprocessing/filtering, then flagship for final output
Many flagship models support system prompts — invest time in prompt engineering to maximize value
Compare Flagship Models
Frequently Asked Questions
What makes a model "flagship"?
Flagship models represent each provider's most capable offering. They excel at complex reasoning, creative tasks, and nuanced understanding. They typically have the largest parameter counts and the most training data.
Are flagship models worth the higher cost?
For tasks requiring deep understanding, creative output, or handling edge cases, yes. For simple classification or extraction, an efficient model often performs equally well at a fraction of the cost.
Which flagship model is the best value?
It depends on your use case. Compare the models on this page using our cost calculator, and check the comparison pages for head-to-head matchups.