[merefield]

Why

OpenAI has published GPT-6.1 Terra (gpt-6.1-terra). Without a maintained price, Codexometer correctly treats its usage as unpriced, preventing useful API-equivalent accounting and benchmark comparisons. This adds the official rates and reviews the existing table against current pricing.

Pricing and audit

Sources retrieved 27 October 2026:

• Official API pricing
• GPT-6.1 Terra model and pricing notes

USD per million tokens:

GPT-6.1 Terra	Input	Cached input	Cache writes (30 minutes)	Output
Standard, up to 272,000 input tokens	$1.00	$0.05	$1.25	$5.00
Standard, above 272,000 input tokens	$2.00	$0.10	$2.50	$7.50

• Add published Fast/priority 2× and Flex 0.5× multipliers for Terra. No Ultrafast rate is guessed.
• Verify all 12 existing maintained models: GPT-6 Astra, GPT-6.1 Sol, GPT-6 Sol/Luna, GPT-5.6 Sol/Terra/Luna, GPT-5.5, GPT-5.4, GPT-5.4 Mini, GPT-5.3 Codex and GPT-5.2. Existing maintained Standard, long-context and service-tier rates remain correct and unchanged.
• Update both embedded pricing-review dates to 2026-10-27, used by the existing TUI/web pricing metadata.
• Use the shared Go estimator; no separate TUI/web pricing implementations or frontend changes.
• Retain fail-closed handling for unpublished gpt-6-terra, preview/provider aliases, unsupported tiers and missing price classes. Benchmark API EQ remains Standard pricing.

Documentation

• Expand the README from the September additions to the full 13-model maintained price table, including GPT-5.2 which was previously omitted from its prose list.
• Correct stale review dates and describe long-context boundaries, the GPT-5.5 Fast exception, and supported tier behaviour.
• State the existing cache-write limitation: estimates use published 30-minute rates; telemetry does not identify retention, so higher-priced 12-hour writes are not separately accounted for.
• Note the currently advertised GPT-5.6 Sol promotional-pricing horizon (at least 21 November 2026).

Validation

• Pin all 13 maintained Standard/long-context price entries and all currently maintained tier multipliers.
• Exercise Terra and existing GPT-6 models at 272,000 / 272,001 input tokens, including cached input, cache writes and output, across Standard, Fast/priority and Flex.
• Check dated/case-normalised model names and rejection of unverified IDs and Terra Ultrafast pricing.
• Refresh the English snapshot solely for the visible pricing-date change.
• Go core suite with race detection; all-language screen regressions; go vet; formatting and diff checks.
• Local Linux and Windows amd64 builds.
• Focused pricing regressions also executed natively on Windows: PASS.

No dependencies, version bump, live model requests or subscription-quota changes.

[Copilot]

🟡 Changes recommended

The README's tier-pricing paragraph still shows the old 12 October 2026 review date and leaves GPT-6.1 Terra out of the Fast/priority and Flex model lists, which contradicts the code and the PR description.

1 open issue

• Update tier-pricing documentation for GPT-6.1 Terra · New

What changed in this PR

This PR adds published pricing for GPT-6.1 Terra (gpt-6.1-terra) to Codexometer. Without a maintained price, Terra usage shows as unpriced, so API-equivalent cost and benchmark comparisons are incomplete. The PR also re-checks the 12 existing maintained models and moves the pricing review dates to 27 October 2026.

Changes:

• Adds Terra Standard rates (up to and above 272,000 input tokens) to the shared price table, plus Fast/priority (2×) and Flex (0.5×) multipliers. No Ultrafast rate is added.
• Updates the embedded pricing review dates and the English snapshot hash to match.
• Expands the README price table to all 13 maintained models and documents long-context, tier, and cache-write behaviour.

Changes:

• Adds Terra pricing and tier multipliers in code.
• Adds tests for Terra and the existing GPT-6 models, including the 272,000 / 272,001 token boundary and rejection of unverified model IDs.

File	Description
README.md	Expands the maintained price table to 13 models, updates the review date, and documents long-context, tier, and cache-write limits.
internal/​ui/​english_snapshot_test.go	Updates the expected snapshot hash for the new pricing review date.
internal/​codex/​service_tier.go	Adds Terra to the Fast/priority (2×) and Flex (0.5×) multiplier lists and updates the Fast pricing review date.
internal/​codex/​service_tier_test.go	Adds tests for Terra tier multipliers, unverified model IDs, and a full multiplier table for each maintained model.
internal/​codex/​benchmark.go	Adds Terra Standard and long-context rates to the price table and updates the standard pricing review date.
internal/​codex/​benchmark_test.go	Pins Terra's published rates and extends boundary tests to Terra, Flex, and rejected model IDs.

_{Review effort: Lite}