fix(pricing): add promo pricing for GPT-5.6 Sol on Bedrock

[lmcdonald-godaddy]

TLDR

Problem this solves:

Bedrock GPT-5.6 Sol still billed pre-Aug 21 rates ($5.50/$33 Geo, $5/$30 Global)

How it solves it:

Applies the AWS promo to Mantle and Converse Sol entries and pins restore-on-2026-11-21 in tests

User Flow

Before: headers come back with x-litellm-response-cost at $5.50 in / $33 out per 1M

After: headers come back with x-litellm-response-cost at $4.40 in / $22 out per 1M (Geo/Mantle). Global CRIS is $4 / $20

Relevant issues

Follow-up to #37880

Linear ticket

Pre-Submission checklist

• 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)

Screenshots / Proof of Fix

Announcement (Aug 21): https://aws.amazon.com/about-aws/whats-new/2026/08/bedrock-openai-gpt-56-sol-reduced-pricing/

Live rates (Global $4 / $20, Geo/In-Region $4.40 / $22): https://docs.aws.amazon.com/bedrock/latest/userguide/model-card-openai-gpt-56-sol.html

Type

Bug Fix

Caveats (if any)

Promo runs at least through 2026-11-21; tests name the sticker rates to restore

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

[greptile-apps]

Greptile Summary

Updates Bedrock GPT-5.6 Sol catalog entries to AWS promotional rates and synchronizes regression expectations across Converse and Mantle pricing tests

• Applies Geo and Mantle input/output/cache pricing of $4.40/$22 per million base tokens
• Applies Global CRIS pricing of $4/$20 per million base tokens
• Updates above-272k, cache-read, and cache-creation rates in both primary and packaged backup catalogs

Confidence Score: 4/5

The pricing update appears safe to merge, with only non-blocking cleanup needed for duplicated maintenance comments

The production and backup catalogs remain synchronized, and regression tests exercise the updated promotional dimensions; the remaining concern is maintainability

Files Needing Attention: tests/test_litellm/llms/bedrock/test_cross_region_inference_profile_mapping.py and tests/test_litellm/llms/bedrock_mantle/test_bedrock_mantle_responses_transformation.py

Important Files Changed

Filename	Overview
model_prices_and_context_window.json	Updates all eight pricing dimensions for the Mantle, US Geo, and Global GPT-5.6 Sol records consistently
litellm/model_prices_and_context_window_backup.json	Keeps the packaged fallback cost map synchronized with the primary pricing catalog
tests/test_litellm/llms/bedrock/test_cross_region_inference_profile_mapping.py	Updates Geo and Global regression expectations but adds verbose duplicated maintenance comments
tests/test_litellm/llms/bedrock_mantle/test_bedrock_mantle_responses_transformation.py	Updates Mantle pricing assertions while repeating the promotional restoration schedule in source comments

_{Reviews (1): Last reviewed commit: "Add promo pricing for gpt 5.6 sol" | Re-trigger Greptile}

[greptile-apps]

Duplicated pricing maintenance comments

These comments duplicate restoration rates across tests and catalogs, increasing synchronization work and allowing future pricing updates to drift

Context Used: CLAUDE.md (source)

Note: If this suggestion doesn't match your team's coding style, reply to this and let me know. I'll remember it for next time!

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[codspeed-hq]

Merging this PR will not alter performance

✅ 31 untouched benchmarks

---

_{Comparing lmcdonald-godaddy:gpt-5.6-sol-promo-pricing (6dbee8c) with litellm_internal_staging (d447be1)¹}

Footnotes

1. No successful run was found on litellm_internal_staging (f818a48) during the generation of this report, so d447be1 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

[devin-ai-integration]

Heads up: #38207 reached the same Sol rates from the same model card, plus a Cyber entry. Either one covers the Sol pricing.

[lmcdonald-godaddy]

Closing as superseded by #38207