AI model API prices: input and output tokens
Input and output tokens are billed separately. This table shows only active, primary, standard base-text API offers with both prices from the same service. Batch, cache, long-context and audio/image-token rates are excluded.
These are provider list prices in USD per 1 million tokens, not a total-cost or quality ranking. Taxes, minimum charges, region, quotas and changes after the check date can affect a real bill. Missing prices are not zero; see each card for other tiers.
Estimate the bill: two worked examples
GPT-6.1 Terra's published OpenAI Standard rates in USD per million tokens, checked 2026-11-03. These are arithmetic examples, not executed API calls. ChatGPT/Copilot subscriptions are separate.
1. Short requests
1,000 requests × (2,000 input + 500 billable output tokens), no cache: 2 million input and 0.5 million output. 2 × $1 + 0.5 × $5 = $4.50. Input includes all submitted context, not only the latest message.
2. Long context
10 requests × (400,000 input + 20,000 billable output tokens), no cache. Each exceeds 272,000 input tokens, so the whole request uses input ×2 and output ×1.5. 4M × $2/M + 0.2M × $7.50/M = $9.50. Applying the multiplier only to excess tokens would be wrong.
Cache, processing tiers and excluded costs
Base cache read: $0.05/M; 30-minute write: $1.25/M; 12-hour write: $2.50/M. Long-context input/cache rates double. Do not count a token as ordinary input, cache read and cache write simultaneously: use its actual usage category.
Batch/Flex is 50% of Standard, Fast is ×2; eligible regional processing adds 10%. Neither example applies these tiers. Totals exclude taxes, paid tools/search, storage, retries, subscriptions and provider markups. Billable output must include applicable reasoning tokens, not just visible text. Check current terms before paying.
| Model | API provider | Input / 1M | Output / 1M | Conditions | Checked | Source |
|---|---|---|---|---|---|---|
| GPT-6 Luna | OpenAI · API | $0.1000 | $0.5000 | Standard; USD per 1M tokens; input context <=272000; original conditions and applicability period retained | 2026-09-26 / 2026-09-26 | Input / 1M / Output / 1M |
| GPT-4o Mini | OpenAI API | $0.1500 | $0.6000 | USD per 1M input tokens; API remains available until January 14, 2027. / USD per 1M output tokens; API remains available until January 14, 2027. | 2026-10-16 / 2026-10-16 | Input / 1M / Output / 1M |
| GPT-5.6 Luna | OpenAI API | $0.2000 | $1.2000 | OpenAI API, standard text pricing | 2026-09-21 / 2026-09-21 | Input / 1M / Output / 1M |
| GPT-5.4 nano | OpenAI · API | $0.2000 | $1.2500 | Standard; USD per 1M tokens; short context only | 2026-09-27 / 2026-09-27 | Input / 1M / Output / 1M |
| Gemini 3.1 Flash Lite Image | Google · Gemini API | $0.2500 | $1.5000 | Standard · text/image/video input; USD per 1M tokens / Standard · text and thinking output; USD per 1M tokens | 2026-09-27 / 2026-09-27 | Input / 1M / Output / 1M |
| GPT-4.1 Mini | OpenAI API | $0.4000 | $1.6000 | USD per 1M input tokens; API remains available until January 14, 2027. / USD per 1M output tokens; API remains available until January 14, 2027. | 2026-10-16 / 2026-10-16 | Input / 1M / Output / 1M |
| Mistral Large 4 | Mistral AI API | $0.7000 | $2.1000 | Standard; USD per 1M input tokens. · Source conditions (original): Стандартный тариф; USD за 1 млн входных токенов. / Standard; USD per 1M output tokens. · Source conditions (original): Стандартный тариф; USD за 1 млн выходных токенов. | 2026-10-30 / 2026-10-30 | Input / 1M / Output / 1M |
| GPT-5.4 Mini | OpenAI · API | $0.7500 | $4.5000 | Standard; USD per 1M tokens; short context only | 2026-09-27 / 2026-09-27 | Input / 1M / Output / 1M |
| GPT-6.1 Terra | OpenAI · API | $1 | $5 | USD per 1,000,000 tokens. OpenAI Standard. For prompts >272,000 input tokens, the entire request uses 2x input/cache rates and 1.5x output. Batch/Flex: 50% below Standard; Fast: 2x Standard; eligible regional processing: +10%. Tool call charges and subscriptions are separate. | 2026-10-27 / 2026-10-27 | Input / 1M / Output / 1M |
| GPT-5.1 | OpenAI · API | $1.2500 | $10 | Standard; USD per 1M tokens; short context only | 2026-09-27 / 2026-09-27 | Input / 1M / Output / 1M |
| GPT-5.2 | OpenAI · API | $1.7500 | $14 | Standard; USD per 1M tokens; short context only | 2026-09-27 / 2026-09-27 | Input / 1M / Output / 1M |
| GPT-5.3-Codex | OpenAI · API | $1.7500 | $14 | Standard; USD per 1M tokens; short context only | 2026-09-27 / 2026-09-27 | Input / 1M / Output / 1M |
| Qwen3.8 Max (0902) | Alibaba Cloud / Qwen Team API | $2 | $6 | Alibaba Cloud International standard pricing | 2026-09-21 / 2026-09-21 | Input / 1M / Output / 1M |
| Qwen3.8-Max | Alibaba Cloud / Qwen Team API | $2 | $6 | Alibaba Cloud international standard scope; regional Global endpoints can differ | 2026-09-21 / 2026-09-21 | Input / 1M / Output / 1M |
| GPT-4.1 | OpenAI API | $2 | $8 | USD per 1M input tokens; API remains available until January 14, 2027. / USD per 1M output tokens; API remains available until January 14, 2027. | 2026-10-16 / 2026-10-16 | Input / 1M / Output / 1M |
| GPT-6 Sol | OpenAI · API | $2 | $10 | Standard; USD per 1M tokens; input context <=272000; original conditions and applicability period retained | 2026-09-26 / 2026-09-26 | Input / 1M / Output / 1M |
| Gemini 4 Argon | Gemini API | $2 | $10 | gemini-4-argon; Standard introductory rate through 2026-11-30 inclusive; USD per 1M tokens. From 2026-12-01: USD 4 per 1M tokens. / gemini-4-argon; Standard introductory rate through 2026-11-30 inclusive; USD per 1M tokens. From 2026-12-01: USD 20 per 1M tokens. | 2026-10-13 / 2026-10-13 | Input / 1M / Output / 1M |
| GPT-5.6 Terra | OpenAI API | $2 | $12 | OpenAI API, standard text pricing | 2026-09-21 / 2026-09-21 | Input / 1M / Output / 1M |
| GPT-4o | OpenAI API | $2.5000 | $10 | USD per 1M input tokens; API remains available until January 14, 2027. / USD per 1M output tokens; API remains available until January 14, 2027. | 2026-10-16 / 2026-10-16 | Input / 1M / Output / 1M |
| GPT-5.4 | OpenAI · API | $2.5000 | $15 | Standard; USD per 1M tokens; input context <272K | 2026-09-27 / 2026-09-27 | Input / 1M / Output / 1M |
| GPT-5.6 Sol | OpenAI API | $4 | $20 | OpenAI API, standard text pricing | 2026-09-21 / 2026-09-21 | Input / 1M / Output / 1M |
| Muse Wonder | Meta Model API | $4 | $20 | muse-wonder-1.0; Standard mode; USD per 1M tokens; geographic restrictions apply. | 2026-10-13 / 2026-10-13 | Input / 1M / Output / 1M |
| GPT-5.5 | OpenAI · API | $5 | $30 | Standard; USD per 1M tokens; input context <=272000; original conditions and applicability period retained | 2026-09-26 / 2026-09-26 | Input / 1M / Output / 1M |
| Claude Fable 5.1 | Anthropic API | $10 | $50 | Claude API standard pricing | 2026-09-21 / 2026-09-21 | Input / 1M / Output / 1M |
| GPT-6 Astra | OpenAI API | $10 | $50 | OpenAI API, standard text pricing; >272K prompt has long-context multiplier | 2026-09-21 / 2026-09-21 | Input / 1M / Output / 1M |
| GPT-5 Pro | OpenAI · API | $15 | $120 | Standard; USD per 1M tokens; short context only | 2026-09-27 / 2026-09-27 | Input / 1M / Output / 1M |
| GPT-5.2 Pro | OpenAI · API | $21 | $168 | Standard; USD per 1M tokens; short context only | 2026-09-27 / 2026-09-27 | Input / 1M / Output / 1M |
| GPT-5.4 Pro | OpenAI · API | $30 | $180 | Standard; USD per 1M tokens; input context <272K | 2026-09-27 / 2026-09-27 | Input / 1M / Output / 1M |
| GPT-5.5 Pro | OpenAI · API | $30 | $180 | Standard; USD per 1M tokens; input context <272K | 2026-09-27 / 2026-09-27 | Input / 1M / Output / 1M |
Related catalog pages
Context comparison · API pricing · Models for programming · Models with API access · Long-context models · Sources and methodology