New model support for Opus 5 and Gemini 3.6 Flash, Greenhouse Recruiting app update, and lower credit costs for Open AI models

here's a quick look at the latest product improvements and fixes claude opus 5 is available in make use claude opus 5 in scenarios that need to reason over complete codebases, detailed specifications, compliance manuals, diagrams, or spreadsheets key benefits more source material in one call review a repository alongside its specification, or assess a complete policy manual, without dividing the material across separate ai handoffs thinking effort you control extended thinking is enabled by default choose low, medium, or high effort to match the depth of reasoning your scenario requires agent coordination build manager agent scenarios in which claude opus 5 delegates work to make ai sub agents, runs writer and verifier loops, and checks results before the next step code and visual analysis review multi file refactoring work, security and architecture requirements, charts, diagrams, and complex spreadsheets for design to code and design qa workflows flexible use use claude opus 5 in make ai toolkit, the anthropic claude app, and make ai agents flexible access you can automate with claude opus 5 using your anthropic api key or make ai provider in make ai provider, select the model from the dropdown with no anthropic account or api key required, and usage is billed through make credits availability available on all make plans gemini 3 6 flash in make use gemini 3 6 flash in to analyze large text and mixed media inputs, then return structured data for the next step in your automation key benefits large source analysis with up to one million input tokens per request, works with extensive transcripts, document collections, or support histories with less prompt splitting mixed media inputs analyzes text, images, audio, or video, so your scenario captures information in any format structured outputs on request, returns json that follows your schema, so you can map fields such as invoice totals, request type, language, and priority into a database, router, or notification module frequent processing runs recurring classification, extraction, and content workflows with input pricing starting at $1 50 per 1 million tokens and output pricing starting at $7 50 per 1 million tokens automation ready use cases extracts decisions and owners from customer calls, routes support requests using account context, or captures invoice details and line items for finance records availability gemini 3 6 flash is available with a gemini api key in make ai toolkit , google gemini ai , and make ai agent (new) apps greenhouse recruiting new app version with harvest api v3 support greenhouse recruiting has released a new app version built on harvest api v3 this replaces the deprecated harvest v1/v2 api and its api key authentication with an oauth connection what changed the new version connects to greenhouse recruiting using oauth instead of an api key several modules also have updated inputs and outputs to match the harvest v3 data structure many nested objects (such as source, stage, or recruiter details) are now flattened into id fields, and some fields, like attachment content type and activity feed notes, were removed greenhouse recruiting is shutting down the harvest v1/v2 api on august 31, 2026 scenarios still using the old app version and api key connection will stop working after that date migrate your scenarios to the new app version and an oauth connection before then to avoid disruption know more see the full module by module migration guide, including field mapping and breaking changes, in the greenhouse recruiting https //apps make com/greenhouse recruiting page lower credit costs for gpt 5 6 terra and luna in openai openai lowered pricing for gpt 5 6 terra by 20% and gpt 5 6 luna by 80% across all rates what changed the simple text prompt module now uses fewer credits per token when you select gpt 5 6 terra or gpt 5 6 luna the update covers both short context runs (up to 272,000 tokens) and long context runs (over 272,000 tokens) gpt 5 6 sol pricing is unchanged existing scenarios that use gpt 5 6 terra or gpt 5 6 luna aren't affected beyond consuming fewer credits per run know more to learn more about credit usage in the simple text prompt module, see openai modules https //apps make com/openai modules