Official API token rates, checked 20 Nov 2026
Standard input / output per million tokens for current OpenAI, Claude, Gemini, and Grok SKUs. Cache, batch, and dated intro rates sit in the notes. Open the linked page before you pay.
Query
Current Standard list price, USD per 1 million tokens, short context. This is the overflow / automation meter. Personal chat and coding stay on the subscription: personal list prices and subscription vs API.
Result
Checked 20 November 2026. Tax extra. Output on Gemini includes thinking tokens.
| Lab | Model | Input | Output | Cached input | Dated change |
|---|---|---|---|---|---|
| OpenAI | gpt-6-luna | $0.10 | $0.50 | $0.01 | Long context (>272K in): $0.20 / $0.75 |
| OpenAI | gpt-6.1-terra | $1.00 | $5.00 | $0.05 | Long: $2.00 / $7.50 |
| OpenAI | gpt-6.2-sol | $2.00 | $10.00 | $0.10 | Long: $4.00 / $15.00 |
| OpenAI | gpt-6.2-astra | $8.00 | $40.00 | $0.40 | Long: $16.00 / $60.00 |
| OpenAI | gpt-5.6-sol | $4.00 | $20.00 | $0.40 | Promo through 30 Nov 2026 |
| Anthropic | Claude Haiku 5.5 (≤100K prompt) | $0.10 | $0.50 | $0.01 | Prompts >100K: $0.50 / $2.50, cache $0.05 |
| Anthropic | Claude Sonnet 5.5 | $2.00 | $10.00 | $0.10 | Cache write (5 min) $2.50 |
| Anthropic | Claude Opus 5.5 | $4.00 | $20.00 | $0.20 | Fast mode 2× |
| Anthropic | Claude Fable 5.2 | $8.00 | $40.00 | $0.20 | Cache write (5 min) $10 |
| Gemini 3.8 Flash | $0.75 | $3.75 | $0.075 | $1.50 / $7.50 from 1 Jan 2027 | |
| Gemini 4 Hydrogen | $1.50 | $7.50 | $0.075 | Free tier exists; Free content trains Google products | |
| Gemini 4 Carbon / Argon | $2.00 | $10.00 | $0.10 | $4.00 / $20.00 from 1 Dec 2026 | |
| Gemini 4 Neon | $3.00 | $15.00 | $0.15 | No Free tier | |
| xAI | grok-4.3 (<200K prompt) | $1.25 | $2.50 | $0.20 | ≥200K: $2.50 / $5.00, cache $0.40 for the whole request |
| xAI | grok-4.7 (<200K prompt) | $2.00 | $6.00 | $0.50 | ≥200K: $4.00 / $12.00, cache $1.00 for the whole request. Lab default for code and chat |
Batch and speed
| Lab | Batch | Other official levers |
|---|---|---|
| OpenAI | 50% of Standard | Flex ~50% on older SKUs. Fast 2× on GPT-6.2. Ultrafast on 6.2 is 3× Standard through 28 Feb 2027, then 6×. Regional / FedRAMP +10% on models from 5 Mar 2026 |
| Anthropic | 50% | Fast mode 2× on Opus 5.5 and Fable 5.2. US-only inference 1.1× |
| 50% | Flex matches Batch on Carbon/Argon/Hydrogen. Priority ~1.8× Standard | |
| xAI | 20% off | Long-prompt band is not optional once the prompt hits 200K |
Last updated
20 November 2026. Re-open the four source URLs in your own account region. Cloud list prices on Azure, Bedrock, and Vertex are set by the cloud, not by this table.
Sources and scope
Personal Standard API. No embeddings, image, audio, or fine-tune rows. No aggregator or relay quotes.
Last checked 2026-11-20 against the links below. Open the vendor page in your region before you pay — list prices move.
- Official · OpenAI API pricing · Open source · Standard short context ≤272K. Long context is a separate column. GPT-5.6 Sol promo through 30 Nov 2026.
- Official · Claude pricing — API · Open source · Platform rates on the consumer pricing page. Batch 50% off. US-only inference 1.1×. Fast mode 2× on Opus 5.5 and Fable 5.2.
- Official · Gemini Developer API pricing · Open source · Carbon/Argon Standard $2/$10 through 30 Nov 2026, $4/$20 from 1 Dec. 3.8 Flash $0.75/$3.75 through 31 Dec 2026.
- Official · xAI API pricing · Open source · Checked 2026-11-21. Prompt ≥200K bills the whole request at the higher band. Batch 20% off.