chore(prices): sync OpenAI prices: 3 models

[berriai-litellm-provider-info-sync]

TLDR

Problem this solves:

• Batch cached input prices for GPT-5 were missing
• Cost tracking for those calls was incomplete

How it solves it:

• Adds cache_read_input_token_cost_batches for three OpenAI models
• Values read from OpenAI's published pricing page

User Flow

This does not apply to an automated price sync; the only change is catalog pricing data

Relevant issues

Affected release

Linear ticket

Pre-Submission checklist

Please complete all items before asking a LiteLLM maintainer to review your PR

• I have added meaningful tests
• The handful of test files covering my change pass locally, e.g. uv run pytest tests/test_litellm/<your_test_file>.py -v. Leave the suites (make test-unit-*, make test-unit) to CI: it finishes in ~15 minutes where a laptop takes an hour or more
• My PR passes all required CI/CD checks (e.g., lint, schema.d.ts sync check, etc.)
• My PR's scope is as isolated as possible; it only solves 1 specific problem
• I have received a Greptile Confidence Score of at least 4/5 before requesting a maintainer review (Greptile reviews automatically once the PR is opened; only comment @greptileai to re-request a review after pushing changes)

Delays in PR merge?

If you're seeing a delay in your PR being merged, ping the LiteLLM Team on Slack (#pr-review).

Screenshots / Proof of Fix

This does not apply to an automated price sync; one of the three models was verified through LiteLLM with a /v1/chat/completions call before the values were applied

Type

Refactoring

Caveats (if any)

Medium

• Six OpenAI rows skipped, parser could not read them
• All six state long-context prices with no threshold
• A human should check those rows against the page

Low

• Four priced rows have no litellm key or field
• Those were deliberately left out, not missed

QA runbook

This does not apply; no tests or e2e files are touched

Final Attestation

• The tests check the right things, including the edge cases, and regressions in the respective real-world customer use-cases are not possible after this PR

Sync report

Updates 3 models, each read from its provider's own published prices.

OpenAI

Updates 3 models from the provider's pricing page (page 3046f29bd4b5).

Verified through LiteLLM before being added: 1 via /v1/chat/completions.

Changes

• gpt-5-2025-08-07: cache_read_input_token_cost_batches $0.0625/1M
• gpt-5-mini-2025-08-07: cache_read_input_token_cost_batches $0.0125/1M
• gpt-5-nano-2025-08-07: cache_read_input_token_cost_batches $0.0025/1M

Skipped by the parser (6)

These rows price models litellm has keys for, but the parser could not read them. A human needs to check them against the page.

• gpt-6-astra: long-context prices present but the row states no threshold
• gpt-6-sol: long-context prices present but the row states no threshold
• gpt-6-luna: long-context prices present but the row states no threshold
• gpt-5.6-sol: long-context prices present but the row states no threshold
• gpt-5.6-terra: long-context prices present but the row states no threshold
• gpt-5.6-luna: long-context prices present but the row states no threshold

No litellm key or field (4)

Rows the page prices that litellm has no key or field for. Nothing to apply; listed so the omission is deliberate, not missed.

• sora-2-pro 1024p: size variant has no litellm key stated on the page
• sora-2-pro 1080p: size variant has no litellm key stated on the page
• ft:o4-mini-2025-04-16: data-sharing discount has no litellm key
• Videos API: no model id in shutdown row | 2026-09-24 | Videos API | --- |

Opened by the litellm-providers price sync. Every price is read from the provider's own published source and gated before it is applied; the audit trail for each value is in the portal's sync history.

---

Note

Low Risk
Pricing-catalog JSON only; no runtime logic changes, but incorrect values would affect batch cached-input cost reporting for these three models.

Overview
Adds cache_read_input_token_cost_batches to the OpenAI gpt-5-2025-08-07, gpt-5-mini-2025-08-07, and gpt-5-nano-2025-08-07 entries in model_prices_and_context_window.json and the backup copy under litellm/.

Each value is half the model’s standard cached-read input rate, matching the existing batch input pricing pattern. This fills a gap so LiteLLM can price batch requests that use prompt cache reads for those GPT-5 snapshots.

^{Reviewed by Cursor Bugbot for commit af4efa4. Bugbot is set up for automated code reviews on this repo. Configure here.}

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[greptile-apps]

Confidence Score: 5/5

The PR appears safe to merge because the catalog-only additions are valid, internally consistent, and synchronized across both pricing files.

Summary

This PR synchronizes batch cached-input pricing for three dated OpenAI GPT-5 models.

• Adds cache_read_input_token_cost_batches for GPT-5, GPT-5 Mini, and GPT-5 Nano dated snapshots.
• Keeps the root catalog and packaged backup synchronized.
• Uses values consistent with the corresponding unversioned model entries.

_{Reviews (1) · Last reviewed commit: "chore(prices): sync OpenAI prices: 3 mod..."}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[codspeed]

Merging this PR will not alter performance

✅ 31 untouched benchmarks

---

_{Comparing litellm-providers/price-sync-openai (af4efa4) with main (9be3c1b)¹}

Footnotes

1. No successful run was found on main (666f6b0) during the generation of this report, so 9be3c1b was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩