Every AI price change CostMon tracks that actually moved a bill, cuts and increases both, with the source and what to do about it.https://costmon.comhttps://costmon.com/price-changes#gpt-5-6-luna-terra-price-cuthttps://costmon.com/price-changes#gpt-5-6-luna-terra-price-cutCiting inference-stack efficiency gains, OpenAI dropped Terra to $2/$12 per million tokens and Luna to $0.20/$1.20, while the top-tier Sol model's price stayed flat. (Luna −80%, Terra −20%) — Confirm the lower rate appears on your next invoice, then check whether tasks still pinned to Sol can safely move down to the now much cheaper Luna or Terra.Thu, 30 Jul 2026 00:00:00 GMTOpenAICuthttps://costmon.com/price-changes#gpt-5-5-price-increasehttps://costmon.com/price-changes#gpt-5-5-price-increaseGPT-5.5 launched at roughly double GPT-5.4's per-token rate. OpenAI's own announcement frames this as a deliberate increase, offset by claimed gains in token efficiency per task. ($5 / $30 per M tokens) — Run a side-by-side cost comparison on real workloads before moving production traffic. Confirm the token savings offset the higher headline price for you.Fri, 24 Apr 2026 00:00:00 GMTOpenAIIncreasehttps://costmon.com/price-changes#gemini-3-flash-caching-batch-pricing-2025https://costmon.com/price-changes#gemini-3-flash-caching-batch-pricing-2025Gemini 3 Flash launched at $0.50 per million input tokens and $3 per million output tokens, with context caching standard rather than opt-in, plus a 50% discount for asynchronous Batch API jobs. (up to −90% w/ caching) — Confirm caching is active on your repeated prompts, and route non-interactive jobs through the Batch API to capture the 50% discount.Wed, 17 Dec 2025 00:00:00 GMTGoogle (Gemini)Cuthttps://costmon.com/price-changes#claude-opus-4-5-pricing-cuthttps://costmon.com/price-changes#claude-opus-4-5-pricing-cutOpus 4.5 launched at $5 input / $25 output per million tokens, a cut Anthropic frames as putting frontier-tier capability within reach of more teams. ($5 / $25 per M tokens) — Re-evaluate workloads kept on Sonnet purely for cost. At the new rate, Opus may be affordable for tasks where the extra capability wasn't worth the old premium.Mon, 24 Nov 2025 00:00:00 GMTAnthropicCuthttps://costmon.com/price-changes#gemini-3-pro-launch-pricing-2025https://costmon.com/price-changes#gemini-3-pro-launch-pricing-2025Google's new flagship reasoning and agentic model launched at $2 per million input tokens and $12 per million output tokens for prompts up to 200K tokens. ($2 / $12 per M tokens) — Benchmark real task cost against your current model before switching. The higher output rate needs to be earned back in fewer tokens per task, not assumed.Tue, 18 Nov 2025 00:00:00 GMTGoogle (Gemini)Restructurehttps://costmon.com/price-changes#claude-haiku-4-5-pricinghttps://costmon.com/price-changes#claude-haiku-4-5-pricingThe new small model launched at a flat $1 input / $5 output per million tokens, positioned by Anthropic as Sonnet-4-level coding at a third of the cost. ($1 / $5 per M tokens) — Benchmark current Sonnet-4-class workloads against Haiku 4.5 and shift eligible traffic down a tier, checking quality on your own tasks first.Wed, 15 Oct 2025 00:00:00 GMTAnthropicRestructurehttps://costmon.com/price-changes#gpt-5-launch-pricinghttps://costmon.com/price-changes#gpt-5-launch-pricingGPT-5 launched at $1.25 per million input tokens and $10 per million output tokens, with mini and nano tiers underneath it: one family instead of separate reasoning and non-reasoning models. ($1.25 / $10 per M tokens) — Map existing GPT-4o and o3 call sites to gpt-5, gpt-5-mini, or gpt-5-nano by task difficulty, and re-benchmark spend rather than assuming a like-for-like swap.Thu, 07 Aug 2025 00:00:00 GMTOpenAIRestructurehttps://costmon.com/price-changes#gemini-2-5-flash-stable-repricing-2025https://costmon.com/price-changes#gemini-2-5-flash-stable-repricing-2025Going stable, Google removed 2.5 Flash's separate thinking vs. non-thinking price tiers, raising input from $0.15 to $0.30 per million tokens while cutting output from $3.50 to $2.50. (in +100%, out −29%) — Recompute cost-per-request at the new $0.30/$2.50 blended rate before migrating off the preview model. Input-heavy workloads got pricier even though output-heavy ones got cheaper.Tue, 17 Jun 2025 00:00:00 GMTGoogle (Gemini)Restructurehttps://costmon.com/price-changes#anthropic-token-saving-updates-2025https://costmon.com/price-changes#anthropic-token-saving-updates-2025A new token-efficient tool-use mode cut output tokens consumed on tool-calling workloads, and prompt-cache reads stopped counting against input-token-per-minute rate limits. (up to −70% output tokens) — Turn on token-efficient tool use for tool-calling workloads, and check whether old throttling logic built around the rate limit is still needed.Thu, 13 Mar 2025 00:00:00 GMTAnthropicCuthttps://costmon.com/price-changes#anthropic-message-batches-api-50-offhttps://costmon.com/price-changes#anthropic-message-batches-api-50-offThe Message Batches API processes up to 10,000 queries asynchronously within 24 hours, at half the price of a standard synchronous call. (−50% vs standard API) — Audit API traffic for anything without a real-time latency requirement and move it to the Batches API to cut that portion of the bill in half.Tue, 08 Oct 2024 00:00:00 GMTAnthropicRestructurehttps://costmon.com/price-changes#gpt-4o-half-price-vs-gpt-4-turbohttps://costmon.com/price-changes#gpt-4o-half-price-vs-gpt-4-turboOpenAI's new flagship model replaced GPT-4 Turbo as the default, matching or beating its quality at half the per-token price. (−50% vs GPT-4 Turbo) — Re-point integrations at gpt-4o and re-run your eval suite before rolling it out. A cheaper model at the same price tier can still shift output style.Mon, 13 May 2024 00:00:00 GMTOpenAICut