From 2902fa084c2179236210fbf416caa9fdb72ad459 Mon Sep 17 00:00:00 2001 From: Bai Li Date: Wed, 7 Oct 2026 11:37:02 -0700 Subject: [PATCH] fix(pricing): add Haiku 5.5, Sonnet 5.5 and GPT-6.1 Sol, re-verify every row Haiku 5.5 (released 2026-10-07) is priced at its up-to-100K-prompt tier; prompts over 100K bill 5x and read low, like the other long-context tiers the card does not model. Sonnet 5.5's cache read is the 0.05x rate that took effect with the Haiku 5.5 launch. GPT-6.1 Sol is new on OpenAI's card. Every other row still matches its public list price. Co-Authored-By: Claude Opus 5.5 --- evalboard/lib/pricing.generated.ts | 3 +++ src/coder_eval/pricing.py | 8 ++++++-- 2 files changed, 9 insertions(+), 2 deletions(-) diff --git a/evalboard/lib/pricing.generated.ts b/evalboard/lib/pricing.generated.ts index 977c4af0..4d28eb79 100644 --- a/evalboard/lib/pricing.generated.ts +++ b/evalboard/lib/pricing.generated.ts @@ -21,6 +21,7 @@ export const PRICING: Record = { "claude-haiku-3-5": { inputPerMTok: 0.8, outputPerMTok: 4.0, cacheWritePerMTok: 1.0, cacheReadPerMTok: 0.08 }, "claude-haiku-4-5": { inputPerMTok: 1.0, outputPerMTok: 5.0, cacheWritePerMTok: 1.25, cacheReadPerMTok: 0.1 }, "claude-haiku-4-5-20251001": { inputPerMTok: 1.0, outputPerMTok: 5.0, cacheWritePerMTok: 1.25, cacheReadPerMTok: 0.1 }, + "claude-haiku-5-5": { inputPerMTok: 0.1, outputPerMTok: 0.5, cacheWritePerMTok: 0.125, cacheReadPerMTok: 0.01 }, "claude-opus-4": { inputPerMTok: 15.0, outputPerMTok: 75.0, cacheWritePerMTok: 18.75, cacheReadPerMTok: 1.5 }, "claude-opus-4-1": { inputPerMTok: 15.0, outputPerMTok: 75.0, cacheWritePerMTok: 18.75, cacheReadPerMTok: 1.5 }, "claude-opus-4-20250514": { inputPerMTok: 15.0, outputPerMTok: 75.0, cacheWritePerMTok: 18.75, cacheReadPerMTok: 1.5 }, @@ -36,6 +37,7 @@ export const PRICING: Record = { "claude-sonnet-4-5-20250929": { inputPerMTok: 3.0, outputPerMTok: 15.0, cacheWritePerMTok: 3.75, cacheReadPerMTok: 0.3 }, "claude-sonnet-4-6": { inputPerMTok: 3.0, outputPerMTok: 15.0, cacheWritePerMTok: 3.75, cacheReadPerMTok: 0.3 }, "claude-sonnet-5": { inputPerMTok: 2.0, outputPerMTok: 10.0, cacheWritePerMTok: 2.5, cacheReadPerMTok: 0.2 }, + "claude-sonnet-5-5": { inputPerMTok: 2.0, outputPerMTok: 10.0, cacheWritePerMTok: 2.5, cacheReadPerMTok: 0.1 }, "codex-mini-latest": { inputPerMTok: 1.5, outputPerMTok: 6.0, cacheWritePerMTok: 1.5, cacheReadPerMTok: 0.375 }, "deepseek.v3.2": { inputPerMTok: 0.74, outputPerMTok: 2.22, cacheWritePerMTok: 0.74, cacheReadPerMTok: 0.0 }, "gemini-3-1-pro-preview": { inputPerMTok: 2.0, outputPerMTok: 12.0, cacheWritePerMTok: 0.0, cacheReadPerMTok: 0.2 }, @@ -73,6 +75,7 @@ export const PRICING: Record = { "gpt-6-astra": { inputPerMTok: 10.0, outputPerMTok: 50.0, cacheWritePerMTok: 12.5, cacheReadPerMTok: 1.0 }, "gpt-6-luna": { inputPerMTok: 0.1, outputPerMTok: 0.5, cacheWritePerMTok: 0.125, cacheReadPerMTok: 0.01 }, "gpt-6-sol": { inputPerMTok: 2.0, outputPerMTok: 10.0, cacheWritePerMTok: 2.5, cacheReadPerMTok: 0.2 }, + "gpt-6.1-sol": { inputPerMTok: 2.0, outputPerMTok: 10.0, cacheWritePerMTok: 2.5, cacheReadPerMTok: 0.1 }, "jev-1.13.0": { inputPerMTok: 0.042, outputPerMTok: 0.0, cacheWritePerMTok: 0.042, cacheReadPerMTok: 0.0 }, "jev-latest": { inputPerMTok: 0.042, outputPerMTok: 0.0, cacheWritePerMTok: 0.042, cacheReadPerMTok: 0.0 }, "kimi-k2-7-code": { inputPerMTok: 0.95, outputPerMTok: 4.0, cacheWritePerMTok: 0.0, cacheReadPerMTok: 0.19 }, diff --git a/src/coder_eval/pricing.py b/src/coder_eval/pricing.py index 2a52afba..0b9b1c3e 100644 --- a/src/coder_eval/pricing.py +++ b/src/coder_eval/pricing.py @@ -5,7 +5,7 @@ Sources: https://platform.claude.com/docs/en/about-claude/pricing, https://developers.openai.com/api/docs/pricing, https://ai.google.dev/gemini-api/docs/pricing, and OpenRouter's live -``/api/v1/models`` (every row re-verified 2026-09-03, except the Bedrock +``/api/v1/models`` (every row re-verified 2026-10-07, except the Bedrock open-weight block: AWS publishes no eu-north-1 figures for those three). SSOT for rates on both halves of the repo: ``evalboard/lib/pricing.generated.ts`` is @@ -77,12 +77,15 @@ class ModelPricing: "claude-opus-4-1": ModelPricing(15.0, 75.0, 18.75, 1.50), "claude-opus-4": ModelPricing(15.0, 75.0, 18.75, 1.50), "claude-opus-4-20250514": ModelPricing(15.0, 75.0, 18.75, 1.50), - # $2/$10, NOT the $3/$15 that Sonnet 4.6 and earlier pay. Do not copy the 4.x row onto it. + # $2/$10, NOT Sonnet 4.6's $3/$15. 5.5's cache hits fell to 0.05x input on 2026-10-07. + "claude-sonnet-5-5": ModelPricing(2.0, 10.0, 2.50, 0.10), "claude-sonnet-5": ModelPricing(2.0, 10.0, 2.50, 0.20), "claude-sonnet-4-6": ModelPricing(3.0, 15.0, 3.75, 0.30), "claude-sonnet-4-5": ModelPricing(3.0, 15.0, 3.75, 0.30), "claude-sonnet-4-5-20250929": ModelPricing(3.0, 15.0, 3.75, 0.30), "claude-sonnet-4-20250514": ModelPricing(3.0, 15.0, 3.75, 0.30), + # CAVEAT: the <=100K-prompt tier; a prompt over 100K bills every bucket at 5x, so long contexts read low. + "claude-haiku-5-5": ModelPricing(0.10, 0.50, 0.125, 0.01), # $1/$5. Not $0.80/$4 — those are Haiku 3.5's rates. "claude-haiku-4-5": ModelPricing(1.0, 5.0, 1.25, 0.10), "claude-haiku-4-5-20251001": ModelPricing(1.0, 5.0, 1.25, 0.10), @@ -120,6 +123,7 @@ class ModelPricing: "gpt-5.6-luna": ModelPricing(0.20, 1.20, 0.25, 0.02), # CAVEAT: flat rate; GPT-6's >272K-input tier (2x input / 1.5x output) is not modelled and reads low. "gpt-6-astra": ModelPricing(10.0, 50.0, 12.50, 1.00), + "gpt-6.1-sol": ModelPricing(2.0, 10.0, 2.50, 0.10), "gpt-6-sol": ModelPricing(2.0, 10.0, 2.50, 0.20), "gpt-6-luna": ModelPricing(0.10, 0.50, 0.125, 0.01), # Keyed on the literal ids ListModels returns. No cache-write fee, so