lmprices

OpenAIGPT-5.6 Terra API pricing

GPT-5.6 Terra costs $2.00 per 1M input tokens and $12.00 per 1M output tokens on OpenAI's API, since .

Current price

Rates in effect since (prices added). Every price links to its source.

Input
$2.00
Output
$12.00
Blended 3:1
$4.50
GPT-5.6 Terra current rates per 1M tokens
Rate≤272K tokens>272K tokens
Standard
Input$2.00$4.00
Cached input$0.20$0.40
Cache write (30 min)$2.50$5.00
Cache write (12 hours)$5.00$10.00
Output$12.00$18.00
Batch
Input$1.00$2.00
Cached input$0.10$0.20
Cache write (30 min)$1.25$2.50
Cache write (12 hours)$2.50$5.00
Output$6.00$9.00

Launch price

GPT-5.6 Terra launched at this price on .

Input
$2.50
Output
$15.00
Blended 3:1
$5.625

Price history

InputOutput
$0$5$10$15Aug 2026Sep 2026Oct 2026 Input $2.50, from July 9, 2026Input $2.00, from July 30, 2026Output $15.00, from July 9, 2026Output $12.00, from July 30, 2026Output $12.00Input $2.00 $0$5$10$15Sep 2026 Input $2.50, from July 9, 2026Input $2.00, from July 30, 2026Output $15.00, from July 9, 2026Output $12.00, from July 30, 2026Output $12.00Input $2.00
Headline standard prices per 1M tokens. A filled circle marks a dated change; an open circle, a change first seen between two captures.

Every price event

3 verified events, newest first. 1 price move. Documented since .

  1. Prices added ·

    • Cache write (12 hours), ≤272K tokens added at $5.00
    • Cache write (12 hours), >272K tokens added at $10.00
    • Batch cache write (12 hours), ≤272K tokens added at $2.50
    • Batch cache write (12 hours), >272K tokens added at $5.00
    Rates from October 20, 2026
    Rate≤272K tokens>272K tokens
    Standard
    Input$2.00$4.00
    Cached input$0.20$0.40
    Cache write (30 min)$2.50$5.00
    Cache write (12 hours)$5.00$10.00
    Output$12.00$18.00
    Batch
    Input$1.00$2.00
    Cached input$0.10$0.20
    Cache write (30 min)$1.25$2.50
    Cache write (12 hours)$2.50$5.00
    Output$6.00$9.00

    12-hour cache write added: standard and batch, both context tiers, from the live pricing table (2.5x input, as the changelog states). All earlier rates unchanged; the existing cache_write (cacheTtl "30m") is the default 30-minute write ('30m write' on the page from 2026-10-20). The 2026-10-15 capture (20261015084552) still shows a single 'Cache writes' column; the 2026-10-26 capture (20261026192814) shows '30m write' and '12h write'. The guide says 12-hour retention is a Responses API option; the Batch table prices it anyway.

    • Price: Pricing page (developers.openai.com)
      quote
      Standard and Batch tables, columns: Model | Short context input | output | cached input | 30m write | 12h write | Long context input | output | cached input | 30m write | 12h write. Standard: | gpt-5.6-terra | $2.00 | $12.00 | $0.20 | $2.50 | $5.00 | $4.00 | $18.00 | $0.40 | $5.00 | $10.00 | Batch: | gpt-5.6-terra | $1.00 | $6.00 | $0.10 | $1.25 | $2.50 | $2.00 | $9.00 | $0.20 | $2.50 | $5.00 | Short context: ≤272K input tokens. Long context: >272K input tokens.
    • Price: Documentation (developers.openai.com)
      quote
      For GPT-5.6 and later, cache writes cost 1.25× the standard, uncached input-token rate for the default 30-minute lifetime. In the Responses API, you can choose a 12-hour minimum lifetime at a write rate of 2.5× ordinary input.
    • Date: Changelog (developers.openai.com)
      quote
      October, 2026 / Oct 20 / Feature · API: v1/responses / You can now choose 12-hour prompt caching in the API. ... 12-hour cache writes cost 2.5x the input token price.

  2. Price change ·

    Rates from July 30, 2026
    Rate≤272K tokens>272K tokens
    Standard
    Input$2.00$4.00
    Cached input$0.20$0.40
    Cache write (30 min)$2.50$5.00
    Output$12.00$18.00
    Batch
    Input$1.00$2.00
    Cached input$0.10$0.20
    Cache write (30 min)$1.25$2.50
    Output$6.00$9.00

    The 2026-07-29 10:31 UTC capture still shows the launch prices; the 2026-07-30 19:51 UTC capture shows the new ones.

    • Price: Archived page (developers.openai.com, captured July 30, 2026)
      quote
      Standard: gpt-5.6-terra | $2.00 | $0.20 | $2.50 | $12.00 | $4.00 | $0.40 | $5.00 | $18.00. Batch: gpt-5.6-terra | $1.00 | $0.10 | $1.25 | $6.00 | $2.00 | $0.20 | $2.50 | $9.00
    • Price: Changelog (developers.openai.com)
      quote
      Jul 30 ... Starting July 30, GPT-5.6 Luna costs 80% less, while GPT-5.6 Terra costs 20% less.
    • Price: Archived page (developers.openai.com, captured July 11, 2026)
      quote
      For GPT-5.6 models and later model families, cache writes cost 1.25× the uncached input token rate. ... For GPT-5.6 models and later model families, use prompt_cache_options.ttl to set the minimum lifetime of all breakpoints written by the request. The only supported value is 30m, which is also the default. A cached prefix remains eligible for reuse for at least 30 minutes, but OpenAI may retain it longer.
    • Date: Changelog (developers.openai.com)
      quote
      Jul 30 ... Starting July 30, GPT-5.6 Luna costs 80% less, while GPT-5.6 Terra costs 20% less.

  3. Launch price ·

    Rates from July 9, 2026
    Rate≤272K tokens>272K tokens
    Standard
    Input$2.50$5.00
    Cached input$0.25$0.50
    Cache write (30 min)$3.125$6.25
    Output$15.00$22.50
    Batch
    Input$1.25$2.50
    Cached input$0.125$0.25
    Cache write (30 min)$1.5625$3.125
    Output$7.50$11.25

    Launch-day capture (2026-07-09 17:50 UTC).

    • Price: Archived page (developers.openai.com, captured July 9, 2026)
      quote
      Standard: gpt-5.6-terra | $2.50 | $0.25 | $3.125 | $15.00 | $5.00 | $0.50 | $6.25 | $22.50. Batch: gpt-5.6-terra | $1.25 | $0.125 | $1.5625 | $7.50 | $2.50 | $0.25 | $3.125 | $11.25
    • Price: Archived page (developers.openai.com, captured July 11, 2026)
      quote
      For GPT-5.6 models and later model families, cache writes cost 1.25× the uncached input token rate. ... For GPT-5.6 models and later model families, use prompt_cache_options.ttl to set the minimum lifetime of all breakpoints written by the request. The only supported value is 30m, which is also the default. A cached prefix remains eligible for reuse for at least 30 minutes, but OpenAI may retain it longer.
    • Date: Changelog (developers.openai.com)
      quote
      Jul 9 ... Released the [GPT-5.6 model family](https://developers.openai.com/api/docs/guides/latest-model?model=gpt-5.6), including GPT-5.6 Sol for frontier capability, GPT-5.6 Terra for a balance of intelligence and cost, and GPT-5.6 Luna for efficient, high-volume workloads. The `gpt-5.6` alias routes requests to `gpt-5.6-sol`.

Gaps in this history

None known.

API ids

Checked .