Change log, by what changed

Tool and API changes

The request contract changed: how tools are called, what a parameter does, or what the API accepts. These break code before they change outputs, and they can do both.

Something about the interface moved. A parameter deprecated, a tool behaviour altered, a serving change the provider documented.

11 entries . window ends 2026-09-24

By provider

OpenAI: 5
Anthropic: 2
Google: 4

Recommended action

Other views

Tool and API changes over twelve months

Tracked provider changes, Oct 2025 to Sep 2026

New version
Weight refresh
Deprecation
Pricing, context and tool changes share these three colours and always carry a label.

The same timeline as a list, 11 changes

2026-09-02 OpenAI, Tool change: slow_down, server_is_overloaded
2026-06-29 Anthropic, Tool change: claude-opus-4-6
2026-06-24 Google, Tool change: gemini-3.5-flash
2026-05-29 OpenAI, Tool change: prompt_cache_retention
2026-03-18 Google, Tool change: Built-in Tools and Function Calling Combination, Grounding with Google Maps
2026-03-13 OpenAI, Tool change: gpt-5.4
2026-02-10 OpenAI, Tool change: Responses API
2026-02-03 OpenAI, Tool change: gpt-5.2, gpt-5.2-codex
2026-01-29 Anthropic, Tool change: claude-sonnet-4-5-20250929
2026-01-29 Google, Tool change: gemini-3-pro-preview, gemini-3-flash-preview
2025-12-19 Google, Tool change: total_thought_tokens

Every entry

Tool and API changes, newest first

OpenAITool change2 September 2026

slow_down, server_is_overloaded

OpenAI wrote that a rapid increase in traffic can now return a 429 error with the slow_down code, and that temporary overload can return a 503 error with the server_is_overloaded code.

Rerun the suiteCheck that your client and your eval harness treat these two codes as a signal to back off rather than as failures, then run a batch large enough to exercise that path.

Source: OpenAI, API changelog . verified 2026-09-22

AnthropicTool change29 June 2026

claude-opus-4-6

Anthropic removed fast mode for Claude Opus 4.6. Requests to claude-opus-4-6 with speed set to fast no longer run at fast speed or premium pricing: they run at standard speed, are billed at standard rates, and do not return an error. The usage.speed field on the response reports the speed used, and Anthropic says to migrate to Claude Opus 4.8 to continue using fast mode.

Plan a migrationA request that asks for fast mode on this id still succeeds, so read usage.speed in your own logs to see which speed you are getting, and move to Claude Opus 4.8 if the feature depends on it.

Source: Anthropic, Claude API release notes . verified 2026-09-22

GoogleTool change24 June 2026

gemini-3.5-flash

Google launched public preview support for the Computer Use tool in Gemini 3.5 Flash.

Agent: Future AGI describes scoring an agent at four resolutions, end to end, trajectory, step and turn, writes that a high end to end score can hide bad tool calls, and warns that strong per step scores can still miss that the agent pursued the wrong goal.

Rerun the suiteRun your agent cases with the Computer Use tool enabled on gemini-3.5-flash and keep the recorded tool calls, not only the final answers.

Source: Google, Gemini API changelog . verified 2026-09-22

OpenAITool change29 May 2026

prompt_cache_retention

OpenAI wrote that for organizations without ZDR, its zero data retention setting, prompt_cache_retention now defaults to 24h instead of in_memory.

Rerun the suiteCheck whether your requests set prompt_cache_retention explicitly, and decide whether the new default is the one you want before your next run.

Source: OpenAI, API changelog . verified 2026-09-22

GoogleTool change18 March 2026

Built-in Tools and Function Calling Combination, Grounding with Google Maps

Google released the Built-in Tools and Function Calling Combination feature, which makes it possible to use Gemini's built-in tools alongside custom function calling tools in a single API call. The same changelog entry says Grounding with Google Maps is now supported for Gemini 3 models going forward.

Agent: Mixing built in tools with custom functions changes which tool is chosen and in what order. Arize Phoenix describes reading tool_path, the ordered tool calls with their arguments, and tool_io, what each tool was called with and what it returned, and warns that a correct looking answer can come from a broken path.

Rerun the suiteIf you combine a built in tool with your own functions in one call, record the tool calls for a set of cases before and after, so a changed selection order is visible.

Source: Google, Gemini API changelog . verified 2026-09-22

OpenAITool change13 March 2026

gpt-5.4

OpenAI wrote that it updated its image encoder to fix a small bug with input_image inputs in gpt-5.4. OpenAI describes it as a bug fix and publishes no new model id alongside it.

Rerun the suiteIf a feature sends images to gpt-5.4, run your image cases again after this date and keep both sets of outputs, because the id you call did not change.

Source: OpenAI, API changelog . verified 2026-09-22

OpenAITool change10 February 2026

Responses API

OpenAI launched server side compaction in the Responses API, launched Skills support across local and hosted execution, and released the Hosted Shell tool with support for networking in containers.

Agent: Arize Phoenix writes that end to end evaluation only inspects the two endpoints, the input and the final output, and that the interesting failures happen in the middle where they cannot be seen, naming reasoning, tool selection and decision path. Compaction and a new hosted tool both sit in that middle.

Rerun the suiteIf you turn on server side compaction or adopt the Hosted Shell tool, run a long session case with the spans recorded rather than only the final answer.

Source: OpenAI, API changelog . verified 2026-09-22

OpenAITool change3 February 2026

gpt-5.2, gpt-5.2-codex

OpenAI optimized its inference stack and states that gpt-5.2 and gpt-5.2-codex now run about 40 percent faster with unchanged weights.

Rerun the suiteRe-measure latency for these two model ids and check that any timing based retry or timeout logic still behaves the way you intended.

Source: OpenAI, OpenAI API changelog . verified 2026-09-22

AnthropicTool change29 January 2026

claude-sonnet-4-5-20250929

Anthropic took structured outputs out of beta on the Claude API for Claude Sonnet 4.5, Claude Opus 4.5 and Claude Haiku 4.5. The notes state that the output_format parameter moved to output_config.format, and that structured outputs remain in public beta on Amazon Bedrock and Microsoft Foundry.

Extraction: The parameter that carries your schema moved, so the request changes while the fields you expect do not. DeepEval defines a golden as an input plus the expected result, which becomes a full test case once the application produces the actual output.
Classification: Where labels come back through a structured output, the shape of the response is part of the feature. Galtea recommends a tiered release gate in which blocking failures such as safety or policy violations block a deploy at any rate, while regressions gate on the change in fail rate.

Revisit the promptMove output_format to output_config.format in your request builders and rerun the schema validation cases on both the Claude API and any partner platform you use.

Source: Anthropic, Claude API release notes . verified 2026-09-22

GoogleTool change29 January 2026

gemini-3-pro-preview, gemini-3-flash-preview

Google launched support for the Computer Use tool in gemini-3-pro-preview and gemini-3-flash-preview.

Agent: Arize Phoenix describes reconstructing tool_path, the ordered tool calls with their arguments, and tool_io, what each tool was called with and what it returned, and warns that a correct looking answer can come from a broken path.

Rerun the suiteBoth ids are preview models, so keep the rerun cycle short for anything built on this tool, and record which preview id each run used.

Source: Google, Gemini API changelog . verified 2026-09-22

GoogleTool change19 December 2025

total_thought_tokens

Google renamed the total_reasoning_tokens field to total_thought_tokens, saying the new name better aligns with the concept of thoughts in thinking models. The change affects v1beta.

Rerun the suiteUpdate anything that reads total_reasoning_tokens from a v1beta response, including cost dashboards and eval harness logging, then run a batch and check the field is populated.

Source: Google, Gemini API changelog . verified 2026-09-22

The full change log

Sources

Change-log entries restate provider announcements with sources and dates. Impact notes cite public practitioner reports. The site publishes no measurements of its own on any model.

Questions about tool and api changes

How is a change classified as a tool and api change?

From what the provider wrote, not from a guess about severity. Where a provider states that weights are unchanged, the entry is not classified as a weight change, and where an unchanged model string starts returning something else, it is.

Is this list complete?

It covers the model families we track for customers, inside a twelve-month window, where the announcement is reachable on the provider's own domain. Anything we cannot reach there is not published, however widely it was reported. Send us a provider link and we will add it.

Do you say what this did to real features?

No. The impact notes say what this kind of change typically touches, each citing a published practitioner source. What it did to your feature is a question only a rerun against your own dataset answers.