> ## Documentation Index
> Fetch the complete documentation index at: https://phaseo.app/docs/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> Phaseo Gateway base URL: https://api.phaseo.app/v1. Send `Authorization: Bearer $PHASEO_API_KEY` with every request.
> Model IDs look like `provider/model`, for example `openai/gpt-6-astra`. Do not guess IDs: list them with `GET /v1/models` or check https://phaseo.app/models. `phaseo/free` routes to an available free model and needs no wallet balance.
> For new text integrations prefer `POST /v1/responses`. `POST /v1/chat/completions` (OpenAI format) and `POST /v1/messages` (Anthropic format) suit existing code.
> The full API contract is https://phaseo.app/docs/openapi/v1/openapi.yaml. Official SDKs: `@phaseo/sdk` (TypeScript) and `phaseo` (Python), plus Go, C#, PHP, Ruby and Rust.
> Search these docs over MCP at https://phaseo.app/docs/mcp, and install Phaseo skills with `npx skills add https://phaseo.app/docs`. If a page is wrong or out of date, report it with the MCP `submit_feedback` tool.
> Manage Phaseo accounts with the `phaseo` CLI (`npm install -g @phaseo/cli`, then `phaseo login` or `phaseo login --device-code`), the `/v1` Management API or the MCP server at `https://mcp.phaseo.app/mcp?access=manage`. Do not send users to the dashboard for account tasks. A person is still needed for sign-up, approving a sign-in, paying the first top-up on the checkout link that `phaseo credits add` returns when no card is saved, confirming each saved-card top-up from the emailed link, and account security (password, MFA, passkeys). Create API keys with `phaseo keys create --name <name> --json`; never guess one. Credit top-ups need `phaseo login --scopes billing:write` first. Task map: https://phaseo.app/docs/v1/developers/manage-from-an-agent.md.

# October

> October 2026 product updates and model releases across Phaseo.

**October 29, 2026**

Models

* [Base Labs: Joey-1](https://phaseo.app/models/baselabs/joey-1)
* [Mistral: Mistral Large 4](https://phaseo.app/models/mistral/mistral-large-4.0)

**October 28, 2026**

Product

![The changelog with the Product filter on and the date list beside it](https://mintcdn.com/aistats/vL_q8VpPiT9Pp5aS/images/changelog/2026-10-28--changelog-filters.png?fit=max\&auto=format\&n=vL_q8VpPiT9Pp5aS\&q=85\&s=1675d3bce8691aec91e6c8865644c545)

* The changelog now lists product and API changes alongside model releases, with **Product**, **API & SDKs** and **Models** filters that keep the date list ([#3261](https://github.com/phaseoteam/Phaseo/pull/3261)).
* The preset editor adds **Publish** next to **Save Draft**, and **Versions and labels** gives each label a version picker, so choosing an earlier version rolls back; see [Prompt Templates](https://phaseo.app/docs/v1/guides/prompt-templates.md) ([#3301](https://github.com/phaseoteam/Phaseo/pull/3301)).
* Opening a published preset version shows its models and template messages, the lines that differ from the default version, and a **Run** button to try it ([#3301](https://github.com/phaseoteam/Phaseo/pull/3301)).

API & SDKs

* `/v1/embeddings` returns `400 validation_error` when a structured input field is a nested object or array; before, it returned `502` ([#3296](https://github.com/phaseoteam/Phaseo/pull/3296)).
* When every provider rejects a request with the same `400`, `413` or `422`, the gateway returns that status and the provider's message instead of `502` ([#3296](https://github.com/phaseoteam/Phaseo/pull/3296)).
* API key spend limit `429` errors show dollar amounts, such as `$0.0117 of $0.01`, and monthly limits keep their `Retry-After` header ([#3296](https://github.com/phaseoteam/Phaseo/pull/3296)).

**October 27, 2026**

Product

* The preset editor's **Versions and labels** section points named labels at published versions, so moving a label rolls requests back in one click; see [Prompt Templates](https://phaseo.app/docs/v1/guides/prompt-templates.md) ([#3283](https://github.com/phaseoteam/Phaseo/pull/3283)).
* The preset model picker offers only models the gateway still serves; before, it also listed retired models that requests could not reach ([#3260](https://github.com/phaseoteam/Phaseo/pull/3260)).
* Running a preset in a workspace with no credits says so and links to **Add credits**; before, it showed a generic error ([#3276](https://github.com/phaseoteam/Phaseo/pull/3276)).
* Pages that load the full model catalogue, such as the preset editor, get it from cache; before, each load took about 3 seconds and some failed ([#3278](https://github.com/phaseoteam/Phaseo/pull/3278)).

API & SDKs

* Preset requests can pin a version or label with `prompt.version` or `model: "@slug@label"`, and an unknown one returns `404 preset_version_not_found`; see [Prompt Templates](https://phaseo.app/docs/v1/guides/prompt-templates.md) ([#3281](https://github.com/phaseoteam/Phaseo/pull/3281)).
* `GET /v1/presets/{id}/labels` lists preset labels, and `PUT` and `DELETE /v1/presets/{id}/labels/{label}` move and remove one; see [Set preset label](https://phaseo.app/docs/v1/api-reference/endpoint/presets-label-set.md) ([#3283](https://github.com/phaseoteam/Phaseo/pull/3283)).
* Streamed Chat Completions, Responses and Messages events now carry the Phaseo model ID in `model`; before, some carried the provider's own ID ([#3273](https://github.com/phaseoteam/Phaseo/pull/3273)).
* `GET /oauth/userinfo` returns `picture`, `name` and `given_name` for tokens with the `profile` scope, taken from the Phaseo profile ([#3189](https://github.com/phaseoteam/Phaseo/pull/3189)).

**October 26, 2026**

Product

* The **Weekly tokens** column on the models page shows real totals again; before, it showed 0 for every model ([#3241](https://github.com/phaseoteam/Phaseo/pull/3241)).

Models

![GPT-6.1 Terra model release artwork](https://mintcdn.com/aistats/UXQ4WoINRDSo32GX/images/model-releases/2026-10-26--openai--gpt-6.1-terra.png?fit=max\&auto=format\&n=UXQ4WoINRDSo32GX\&q=85\&s=8cd1431eabf7d94ea61905d246303492)

* [OpenAI: GPT-6.1 Terra](https://phaseo.app/models/openai/gpt-6.1-terra)

**October 24, 2026**

Product

* Every workspace can use retired model routing: owners and admins turn it on in **Settings -> Routing -> Retired model routing**, and requests go to the approved replacement; see [Model Deprecation and Retirement](https://phaseo.app/docs/v1/guides/model-retirement.md) ([#2899](https://github.com/phaseoteam/Phaseo/pull/2899), [#2896](https://github.com/phaseoteam/Phaseo/pull/2896)).
* Retired model routing uses an approved replacement only when it costs no more than the retired model, unless owners and admins allow a higher price; a replacement they choose themselves is used whatever it costs ([#2899](https://github.com/phaseoteam/Phaseo/pull/2899), [#3211](https://github.com/phaseoteam/Phaseo/pull/3211)).
* The presets list and editor open even when the full model catalogue fails to load; before, the page showed an error ([#3221](https://github.com/phaseoteam/Phaseo/pull/3221)).

API & SDKs

* Responses served by a retired model's replacement carry the `x-phaseo-model-substitution` header; see [Model Deprecation and Retirement](https://phaseo.app/docs/v1/guides/model-retirement.md) ([#2896](https://github.com/phaseoteam/Phaseo/pull/2896)).
* The Terraform provider adds `phaseo_rate_limit_policy`, `phaseo_rate_limit_assignment` and `phaseo_department_budget` resources, all importable by ID; see [Rate Limits](https://phaseo.app/docs/v1/api-reference/limits.md) ([#3216](https://github.com/phaseoteam/Phaseo/pull/3216)).

**October 23, 2026**

Product

* Owners and admins can create rate limit policies in **Settings -> Rate limits** with requests per minute, tokens per minute, concurrent requests and **Strict mode**; see [Rate Limits](https://phaseo.app/docs/v1/api-reference/limits.md) ([#3150](https://github.com/phaseoteam/Phaseo/pull/3150), [#3213](https://github.com/phaseoteam/Phaseo/pull/3213)).
* **Settings -> Rate limits** assigns a policy to an API key, member, department, guardrail or the end-user default, and shows its throttled requests over 24 hours ([#3150](https://github.com/phaseoteam/Phaseo/pull/3150)).
* Owners and admins can add daily, weekly or monthly department budgets in **Settings -> Budgets**; they send alerts at 80% and 100% and never block requests ([#3208](https://github.com/phaseoteam/Phaseo/pull/3208)).
* **Key Security** moves to **Settings -> API Keys -> Key Security**, and the keys table adds a **Status** column and sortable columns ([#3184](https://github.com/phaseoteam/Phaseo/pull/3184)).
* Broadcast can send traces to Amazon S3 and S3-compatible storage, Cloudflare R2, Google Cloud Storage, ClickHouse, BigQuery and Snowflake; see [Broadcast](https://phaseo.app/docs/v1/guides/broadcast.md) ([#3181](https://github.com/phaseoteam/Phaseo/pull/3181), [#3182](https://github.com/phaseoteam/Phaseo/pull/3182), [#3185](https://github.com/phaseoteam/Phaseo/pull/3185)).
* New guide: [Broadcast](https://phaseo.app/docs/v1/guides/broadcast.md) lists the settings for every destination type ([#3190](https://github.com/phaseoteam/Phaseo/pull/3190)).
* The compare page adds a 30-day **Activity** chart of each model's tokens and requests through Phaseo ([#3147](https://github.com/phaseoteam/Phaseo/pull/3147)).
* The compare page shows where each model ranks on the capability indexes ([#3148](https://github.com/phaseoteam/Phaseo/pull/3148)).
* Two-model comparisons have their own pages with a summary and FAQ, such as `/compare/anthropic_claude-opus-5.5-vs-openai_gpt-6-astra` ([#3162](https://github.com/phaseoteam/Phaseo/pull/3162)).

API & SDKs

* Rate limit policies enforce `concurrent_requests`, returning `429` with reason `policy_concurrent_requests` and `Retry-After: 1` when every slot is taken; see [Rate Limits](https://phaseo.app/docs/v1/api-reference/limits.md) ([#3212](https://github.com/phaseoteam/Phaseo/pull/3212)).
* Strict rate limit policies (`fail_closed`) check their per-minute limits before the request is sent to a provider ([#3212](https://github.com/phaseoteam/Phaseo/pull/3212)).
* Requests sent through the EU and US regional gateways now reach Broadcast destinations ([#3187](https://github.com/phaseoteam/Phaseo/pull/3187)).
* Requests to most models succeed again; on October 23, after a provider data policy update, they returned `500 gateway_context_failed` ([#3206](https://github.com/phaseoteam/Phaseo/pull/3206)).

Models

![Kimi K3.1 Pro model release artwork](https://mintcdn.com/aistats/UXQ4WoINRDSo32GX/images/model-releases/2026-10-23--moonshotai--kimi-k3.1-pro.png?fit=max\&auto=format\&n=UXQ4WoINRDSo32GX\&q=85\&s=6e7bca5e18c5d4cb54bcdbc93efa7d6f)

* [Moonshot: Kimi K3.1 Pro](https://phaseo.app/models/moonshotai/kimi-k3.1-pro)
* [Abliteration AI: Abliterated Large Flash](https://phaseo.app/models/abliteration-ai/abliterated-model-large-flash)

**October 22, 2026**

Product

* The new **Key Security** page flags API keys with no spend limit, expiry, recent use, IP allowlist or recorded storage location, and suggests a fix ([#3159](https://github.com/phaseoteam/Phaseo/pull/3159)).
* Each API key can record where it is stored, such as Vercel, GitHub Actions or 1Password ([#3158](https://github.com/phaseoteam/Phaseo/pull/3158)).
* **Settings -> Budgets** shows each workspace budget's spend, remaining amount and reset time ([#3095](https://github.com/phaseoteam/Phaseo/pull/3095)).
* Broadcast adds 16 destinations, including Langfuse, Datadog LLM Observability, LangSmith, PostHog and Sentry, each with a test button; see [Broadcast](https://phaseo.app/docs/v1/guides/broadcast.md) ([#2962](https://github.com/phaseoteam/Phaseo/pull/2962), [#3080](https://github.com/phaseoteam/Phaseo/pull/3080), [#3083](https://github.com/phaseoteam/Phaseo/pull/3083)).
* Broadcast settings show whether each destination is delivering ([#3102](https://github.com/phaseoteam/Phaseo/pull/3102)).
* **Settings -> Notifications** can send an alert when a Broadcast destination keeps failing ([#3105](https://github.com/phaseoteam/Phaseo/pull/3105)).
* Broadcast webhook destinations accept an optional signing secret ([#3129](https://github.com/phaseoteam/Phaseo/pull/3129)).
* Request details say when zero data retention skipped storing a request's content ([#3178](https://github.com/phaseoteam/Phaseo/pull/3178)).
* The compare page lets you pick the route and service tier under each model, and its prices, cost calculator and verdicts follow that choice ([#3133](https://github.com/phaseoteam/Phaseo/pull/3133)).

API & SDKs

* `/v1/messages` accepts mid-conversation `role: "system"` messages in the positions Claude allows; see [Create a message](https://phaseo.app/docs/v1/api-reference/endpoint/anthropic-messages.md) ([#3165](https://github.com/phaseoteam/Phaseo/pull/3165)).
* The key endpoints and every SDK accept and return `storage_locations`, up to eight places where a key is kept, such as `["railway", "github"]`; see [Create API key](https://phaseo.app/docs/v1/api-reference/endpoint/keys-create.md) ([#3164](https://github.com/phaseoteam/Phaseo/pull/3164)).
* **Beta:** `/v1/department-budgets` sets daily, weekly or monthly USD budgets per department, which alert owners and admins at about 80% and 100% without blocking requests ([#3125](https://github.com/phaseoteam/Phaseo/pull/3125)).
* Zero data retention keeps a request's prompt and response content out of the response cache, data samples and Phaseo's telemetry, whether set by the workspace, a key guardrail or the request ([#3161](https://github.com/phaseoteam/Phaseo/pull/3161), [#3175](https://github.com/phaseoteam/Phaseo/pull/3175)).
* Custom webhook notification deliveries are signed with `x-phaseo-timestamp` and `x-phaseo-signature` (HMAC-SHA256); see [Workspace Notifications](https://phaseo.app/docs/v1/guides/workspace-notifications.md) ([#3089](https://github.com/phaseoteam/Phaseo/pull/3089)).
* Broadcast exports a request's `metadata` as `phaseo.metadata.<key>` span attributes ([#3127](https://github.com/phaseoteam/Phaseo/pull/3127)).
* `trace_name`, `span_name`, `generation_name` and `trace_id` in a request's `metadata` name and group its exported Broadcast traces ([#3127](https://github.com/phaseoteam/Phaseo/pull/3127)).

**October 21, 2026**

Product

* Logged Chat Completions, Responses and Messages requests open in Chat to continue, from request details, the logs table's right-click menu or a session turn; see [Open Requests in Chat](https://phaseo.app/docs/v1/guides/open-requests-in-chat.md) ([#3084](https://github.com/phaseoteam/Phaseo/pull/3084), [#3092](https://github.com/phaseoteam/Phaseo/pull/3092)).
* Owners and admins can limit stored prompts and responses to owners and admins in **Settings -> Privacy -> Who can view contents** ([#3103](https://github.com/phaseoteam/Phaseo/pull/3103)).
* With **Require zero data retention** on, request and response payloads are never stored in request logs ([#3081](https://github.com/phaseoteam/Phaseo/pull/3081)).
* Presets can hold a prompt template with `{{variable}}` placeholders, so requests send only `prompt.id` and the variables; see [Prompt Templates](https://phaseo.app/docs/v1/guides/prompt-templates.md) ([#3085](https://github.com/phaseoteam/Phaseo/pull/3085)).
* The preset editor previews a prompt template as you edit it and has a **Run** button to test it ([#3086](https://github.com/phaseoteam/Phaseo/pull/3086), [#3122](https://github.com/phaseoteam/Phaseo/pull/3122)).
* The **Trace** tab in request details, showing routing, provider attempts, tool calls and the final response, is available to every workspace ([#3082](https://github.com/phaseoteam/Phaseo/pull/3082)).
* The compare page uses one grouped table that highlights the best value in each row, with a switch to show only the rows that differ ([#3116](https://github.com/phaseoteam/Phaseo/pull/3116)).
* `/models` and `/models/table` add a **License** filter, grouped into Permissive, Restricted and Proprietary, and a **Weights** filter for open or closed weights ([#3062](https://github.com/phaseoteam/Phaseo/pull/3062), [#3075](https://github.com/phaseoteam/Phaseo/pull/3075), [#3079](https://github.com/phaseoteam/Phaseo/pull/3079)).

API & SDKs

* **Breaking:** Error responses return `error` as an object with `code`, `message`, `type`, `param` and `metadata`; code that read `error` as a string should read the new top-level `code`; see [Errors and Debugging](https://phaseo.app/docs/v1/api-reference/errors.md) ([#3055](https://github.com/phaseoteam/Phaseo/pull/3055)).
* Chat Completions, Responses, Messages and embeddings responses include `usage.cost` in USD and a `usage.cost_details` breakdown ([#3050](https://github.com/phaseoteam/Phaseo/pull/3050)).
* `POST /v1/keys` returns the new secret at the top level as `key`; see [Create API key](https://phaseo.app/docs/v1/api-reference/endpoint/keys-create.md) ([#3050](https://github.com/phaseoteam/Phaseo/pull/3050)).
* Text endpoints accept explicit `null` for optional parameters, `service_tier: "auto"` and `"scale"`, and `prediction`, which used to return `400` ([#3049](https://github.com/phaseoteam/Phaseo/pull/3049)).
* The `web` and `web-fetch` plugins work on any model, running as Phaseo's web search and web fetch server tools; see [Web Search Server Tool](https://phaseo.app/docs/v1/guides/server-tools/web-search.md) ([#3051](https://github.com/phaseoteam/Phaseo/pull/3051)).
* Rate limit policies enforce `tokens_per_minute` and return `x-ratelimit-limit-tokens`, `x-ratelimit-remaining-tokens` and `x-ratelimit-reset-tokens` headers; see [Rate Limits](https://phaseo.app/docs/v1/api-reference/limits.md) ([#2961](https://github.com/phaseoteam/Phaseo/pull/2961)).
* Rate limit policies assigned to a department now apply to that department's requests ([#3088](https://github.com/phaseoteam/Phaseo/pull/3088)).
* Prompt cache writes bill at the documented rates: 2x input for 1-hour Claude writes and 2.5x input for 12-hour OpenAI writes; see [Prompt Caching](https://phaseo.app/docs/v1/guides/prompt-caching.md) ([#3078](https://github.com/phaseoteam/Phaseo/pull/3078), [#3076](https://github.com/phaseoteam/Phaseo/pull/3076)).
* Responses usage returns `input_tokens_details.cache_write_tokens_details` with the 30-minute and 12-hour cache write split ([#3078](https://github.com/phaseoteam/Phaseo/pull/3078)).

Models

* [Base Labs: Elephant-1](https://phaseo.app/models/baselabs/elephant-1)
* [Base Labs: Mammoth-1](https://phaseo.app/models/baselabs/mammoth-1)

**October 20, 2026**

Models

* [Mistral: Voxtral Mini Transcribe 3](https://phaseo.app/models/mistral/voxtral-mini-transcribe-3)
* [Mistral: Voxtral Small Transcribe 3](https://phaseo.app/models/mistral/voxtral-small-transcribe-3)
* [Mistral: Voxtral TTS 3](https://phaseo.app/models/mistral/voxtral-tts-3)

**October 19, 2026**

API & SDKs

* Optional parameters the provider doesn't support, such as `top_k` or `min_p`, are dropped and listed in the `X-Phaseo-Dropped-Params` response header; see [Parameters](https://phaseo.app/docs/v1/api-reference/parameters.md) ([#2999](https://github.com/phaseoteam/Phaseo/pull/2999)).
* A provider failure mid-stream ends the stream with the endpoint's error event instead of a normal finish ([#2990](https://github.com/phaseoteam/Phaseo/pull/2990)).
* A request that every provider rejected with `429` returns `429` with the provider's `Retry-After` ([#2990](https://github.com/phaseoteam/Phaseo/pull/2990)).
* A request rejected as invalid no longer falls back to the other models in `models` ([#2990](https://github.com/phaseoteam/Phaseo/pull/2990)).

Models

![MiniMax M3.1 Flash model release artwork](https://mintcdn.com/aistats/UXQ4WoINRDSo32GX/images/model-releases/2026-10-19--minimax--minimax-m3.1-flash.png?fit=max\&auto=format\&n=UXQ4WoINRDSo32GX\&q=85\&s=5867feb70972d5af040b750c8f3882cb)

* [MiniMax: MiniMax M3.1 Flash](https://phaseo.app/models/minimax/minimax-m3.1-flash)
* [Cloudflare: Clairic](https://phaseo.app/models/cloudflare/clairic)
* [Inception: Mercury 3](https://phaseo.app/models/inception/mercury-3)

**October 18, 2026**

Product

* English docs pages shared on social media or in chat show a preview image with the page's title and description ([#3003](https://github.com/phaseoteam/Phaseo/pull/3003)).

API & SDKs

* The Phaseo plugin for Claude Code bundles the Phaseo MCP server, the Docs MCP server at `https://phaseo.app/docs/mcp` and Phaseo's skills; see [CLI and MCP](https://phaseo.app/docs/v1/developers/cli-and-mcp.md) ([#2949](https://github.com/phaseoteam/Phaseo/pull/2949)).
* Prompt caching is billed at each provider's documented rates; see [Prompt Caching](https://phaseo.app/docs/v1/guides/prompt-caching.md) ([#2964](https://github.com/phaseoteam/Phaseo/pull/2964)).
* Agent turns stay on the provider that holds their prompt cache ([#2969](https://github.com/phaseoteam/Phaseo/pull/2969)).
* Claude requests accept an `agent` cache scope that places cache breakpoints for multi-turn agents; see [Prompt Caching](https://phaseo.app/docs/v1/guides/prompt-caching.md) ([#2957](https://github.com/phaseoteam/Phaseo/pull/2957)).

**October 17, 2026**

Product

* **Settings -> Notifications** adds **Rate Limit Alerts**, sent at most hourly by email, Slack, Discord, Teams or webhook; see [Workspace Notifications](https://phaseo.app/docs/v1/guides/workspace-notifications.md) ([#2946](https://github.com/phaseoteam/Phaseo/pull/2946)).

Models

![DeepSeek V4.1 Flash 1017 model release artwork](https://mintcdn.com/aistats/UXQ4WoINRDSo32GX/images/model-releases/2026-10-17--deepseek--deepseek-v4.1-flash-1017.png?fit=max\&auto=format\&n=UXQ4WoINRDSo32GX\&q=85\&s=528ef51e53345875436e9e18d8f59c94)

* [DeepSeek: DeepSeek V4.1 Flash 1017](https://phaseo.app/models/deepseek/deepseek-v4.1-flash-1017)

**October 16, 2026**

Product

* The docs use phaseo.app's design, with a new theme, simpler navigation and a footer with Models, Status and Support links ([#2930](https://github.com/phaseoteam/Phaseo/pull/2930)).
* Every docs page's Markdown and `llms.txt` include instructions for AI agents on authentication, model IDs and endpoints ([#2947](https://github.com/phaseoteam/Phaseo/pull/2947)).
* Phaseo's public skills are published under `/.well-known/agent-skills/` and install with `npx skills add https://phaseo.app/docs` ([#2947](https://github.com/phaseoteam/Phaseo/pull/2947)).

API & SDKs

* The SDK docs add the Java SDK, installed from source; see [Java SDK — Installation](https://phaseo.app/docs/v1/sdk-reference/java/installation.md) ([#2904](https://github.com/phaseoteam/Phaseo/pull/2904)).

**October 15, 2026**

Product

* Signed-in users see a notice under the site header when a model their workspace used in the last 90 days is being deprecated; see [Model Deprecation and Retirement](https://phaseo.app/docs/v1/guides/model-retirement.md) ([#2871](https://github.com/phaseoteam/Phaseo/pull/2871)).
* Model deprecation emails go to the workspace owner, and admins can be added ([#2868](https://github.com/phaseoteam/Phaseo/pull/2868)).
* Usage overview forecasts follow the selected date range ([#2921](https://github.com/phaseoteam/Phaseo/pull/2921)).
* Every figure on the usage overview comes from exact totals; before, it used a sample of recent requests ([#2927](https://github.com/phaseoteam/Phaseo/pull/2927)).

API & SDKs

* **Beta:** `/v1/rate-limits` creates and assigns rate limit policies, and the gateway enforces their `requests_per_minute` for keys, members, guardrails and end users; see [Rate Limits](https://phaseo.app/docs/v1/api-reference/limits.md) ([#2888](https://github.com/phaseoteam/Phaseo/pull/2888), [#2893](https://github.com/phaseoteam/Phaseo/pull/2893)).
* Requests over a policy's `requests_per_minute` return `429 rate_limit_exceeded` with `x-ratelimit-*` headers ([#2893](https://github.com/phaseoteam/Phaseo/pull/2893)).
* Retired models keep serving while any provider route is enabled; once none is left, requests return `410 model_retired`; see [Model Deprecation and Retirement](https://phaseo.app/docs/v1/guides/model-retirement.md) ([#2884](https://github.com/phaseoteam/Phaseo/pull/2884)).
* The Batch API runs Gemini models on Vertex AI with Phaseo-managed keys; see [Video and batch jobs](https://phaseo.app/docs/v1/guides/async-video-and-batch.md) ([#2932](https://github.com/phaseoteam/Phaseo/pull/2932), [#2935](https://github.com/phaseoteam/Phaseo/pull/2935)).
* `/v1/batches/models` lists Gemini batch offers ([#2936](https://github.com/phaseoteam/Phaseo/pull/2936)).

Models

* [Nvidia: Nemotron 3.5 Super VL 120B A12B](https://phaseo.app/models/nvidia/nemotron-3.5-super-vl-120b-a12b)

**October 14, 2026**

Product

* **Settings -> Billing & Credits** shows how long your balance will last at the current pace ([#2898](https://github.com/phaseoteam/Phaseo/pull/2898)).
* API key limits show where each window is heading before it resets ([#2895](https://github.com/phaseoteam/Phaseo/pull/2895)).
* Weekly and monthly API key usage shows real amounts; before, it read as zero ([#2895](https://github.com/phaseoteam/Phaseo/pull/2895)).

API & SDKs

* `GET /v1/models` returns an `ETag` and answers `304 Not Modified` to a matching `If-None-Match`; see [List models](https://phaseo.app/docs/v1/api-reference/endpoint/models.md) ([#2916](https://github.com/phaseoteam/Phaseo/pull/2916)).
* Each provider offer in `GET /v1/models` lists the service tiers it accepts ([#2916](https://github.com/phaseoteam/Phaseo/pull/2916)).

Models

![Jev 1.14 model release artwork](https://mintcdn.com/aistats/UXQ4WoINRDSo32GX/images/model-releases/2026-10-14--typesafe--jev-1.14.0.png?fit=max\&auto=format\&n=UXQ4WoINRDSo32GX\&q=85\&s=62ee6c874a3c5111389793d430daaabb)

* [TypeSafe: Jev 1.14](https://phaseo.app/models/typesafe/jev-1.14.0)
* [SpaceXAI: Grok Voice 3.0](https://phaseo.app/models/spacex-ai/grok-voice-3.0)

**October 13, 2026**

Product

* Each usage overview card shows a month-end forecast for spend, requests and tokens, or the month so far for rates such as cache hit rate ([#2900](https://github.com/phaseoteam/Phaseo/pull/2900)).

Models

![Gemini 4 Argon model release artwork](https://mintcdn.com/aistats/UXQ4WoINRDSo32GX/images/model-releases/2026-10-13--google--gemini-4-argon.png?fit=max\&auto=format\&n=UXQ4WoINRDSo32GX\&q=85\&s=af168e3c48416c019832207b9d62eefd)

* [Google: Gemini 4 Argon](https://phaseo.app/models/google/gemini-4-argon)

**October 12, 2026**

Product

* The models page adds OCR, Document parsing and Translation groups ([#2867](https://github.com/phaseoteam/Phaseo/pull/2867)).
* On the models page, **Audio** opens a row of Speech, Transcription, Translation, Realtime and Music filters ([#2873](https://github.com/phaseoteam/Phaseo/pull/2873), [#2876](https://github.com/phaseoteam/Phaseo/pull/2876)).

API & SDKs

* The top-level `models` fallback list works on every inference endpoint, including `/v1/messages` and the media endpoints; see [Routing and Fallbacks](https://phaseo.app/docs/v1/guides/routing-and-fallbacks.md) ([#2858](https://github.com/phaseoteam/Phaseo/pull/2858)).
* `429` responses for API key request and spend limits include `Retry-After` with the seconds until the limit resets ([#2864](https://github.com/phaseoteam/Phaseo/pull/2864)).

Models

![Muse Wonder 1.0 model release artwork](https://mintcdn.com/aistats/UXQ4WoINRDSo32GX/images/model-releases/2026-10-12--meta--muse-wonder-1.0.png?fit=max\&auto=format\&n=UXQ4WoINRDSo32GX\&q=85\&s=2f12aa9e3915da0d1f8eddeedf98527d)

* [Meta: Muse Wonder 1.0](https://phaseo.app/models/meta/muse-wonder-1.0)

**October 11, 2026**

Product

* Model cards show the cheapest provider's standard input and output prices, with units ([#2831](https://github.com/phaseoteam/Phaseo/pull/2831), [#2832](https://github.com/phaseoteam/Phaseo/pull/2832)).
* The model card pricing hover card names the provider behind the price and lists each priced variant of image and video models ([#2845](https://github.com/phaseoteam/Phaseo/pull/2845), [#2839](https://github.com/phaseoteam/Phaseo/pull/2839)).

API & SDKs

* `GET /v1/models?format=openai` returns the model list in the OpenAI list format, with canonical model IDs; see [List models](https://phaseo.app/docs/v1/api-reference/endpoint/models.md) ([#2838](https://github.com/phaseoteam/Phaseo/pull/2838)).

**October 9, 2026**

Product

* Chat adds an opt-in **Interactive answers** experiment, turned on beside temporary chat, that can answer with comparison cards, tables, bar charts and follow-up forms ([#2778](https://github.com/phaseoteam/Phaseo/pull/2778)).

API & SDKs

* Each workspace member has a requests-per-minute allowance shared across their API keys, rising from 10 to 120 as the workspace becomes established; see [Rate Limits](https://phaseo.app/docs/v1/api-reference/limits.md) ([#2690](https://github.com/phaseoteam/Phaseo/pull/2690), [#2800](https://github.com/phaseoteam/Phaseo/pull/2800)).
* Each user can make 1,500 free-model requests per UTC day, and requests over either limit return `429` with `Retry-After` ([#2690](https://github.com/phaseoteam/Phaseo/pull/2690), [#2787](https://github.com/phaseoteam/Phaseo/pull/2787)).
* The EU and US regional gateways return `403 model_region_unavailable` when no provider serves the model in that region; see [Route text requests by region](https://phaseo.app/docs/v1/guides/regional-routing.md) ([#2811](https://github.com/phaseoteam/Phaseo/pull/2811), [#2813](https://github.com/phaseoteam/Phaseo/pull/2813)).

Models

![Clef Omni model release artwork](https://mintcdn.com/aistats/UXQ4WoINRDSo32GX/images/model-releases/2026-10-09--cloudflare--clef-omni.png?fit=max\&auto=format\&n=UXQ4WoINRDSo32GX\&q=85\&s=a419e8521deff2bc890fceaf6d21f1ce)

* [Cloudflare: Clef Omni](https://phaseo.app/models/cloudflare/clef-omni)

**October 7, 2026**

API & SDKs

* `/v1/decisions` accepts OpenAI-compatible input with ordered questions and answers, and existing request shapes keep working; see [Make a decision](https://phaseo.app/docs/v1/api-reference/endpoint/decisions.md) ([#2744](https://github.com/phaseoteam/Phaseo/pull/2744)).

Models

![Claude Haiku 5.5 model release artwork](https://mintcdn.com/aistats/EzqUOO_knTIHSpCt/images/model-releases/2026-10-07--anthropic--claude-haiku-5.5.png?fit=max\&auto=format\&n=EzqUOO_knTIHSpCt\&q=85\&s=e6b3fba09ce243d283d4b8d76f7de15d)

* [Anthropic: Claude Haiku 5.5](https://phaseo.app/models/anthropic/claude-haiku-5.5)

**October 6, 2026**

Product

* Model pages show saved specifications such as total and active parameter counts, licence and training tokens ([#2728](https://github.com/phaseoteam/Phaseo/pull/2728)).

Models

![Mistral Large 4 model release artwork](https://mintcdn.com/aistats/EzqUOO_knTIHSpCt/images/model-releases/2026-10-06--mistral--mistral-large-4.0.png?fit=max\&auto=format\&n=EzqUOO_knTIHSpCt\&q=85\&s=5d8c936ba500efa1d3cee74452d21d3c)

* [Mistral: Mistral Large 4](https://phaseo.app/models/mistral/mistral-large-4.0)
* [Google: Nano Banana 2.1](https://phaseo.app/models/google/gemini-nano-banana-2.1)
* [Google: EmbeddingGemma 2](https://phaseo.app/models/google/embeddinggemma-2)

**October 4, 2026**

API & SDKs

* Every request gets a server-generated `G` generation ID shared by response headers, payloads and logs, and a caller's `x-request-id` no longer replaces it; see [Get a generation](https://phaseo.app/docs/v1/api-reference/endpoint/generation.md) ([#2710](https://github.com/phaseoteam/Phaseo/pull/2710)).

**October 3, 2026**

Product

* Owners and admins can open a workspace member's profile from their user chip, with 30 days of requests and spend, top models, keys and recent logs ([#2700](https://github.com/phaseoteam/Phaseo/pull/2700)).

**October 2, 2026**

Product

* Each API key has its own page with usage charts, limits and settings ([#2686](https://github.com/phaseoteam/Phaseo/pull/2686)).
* API keys can be restricted to labelled IPv4 or IPv6 addresses and CIDR ranges ([#2686](https://github.com/phaseoteam/Phaseo/pull/2686)).
* The web app and docs are translated into Spanish, French, German, Brazilian Portuguese, Japanese, Simplified Chinese, Hindi and Arabic, including text that used to fall back to English ([#2681](https://github.com/phaseoteam/Phaseo/pull/2681)).
* The public Monitor page is removed, and saved Monitor landing preferences open Home ([#2683](https://github.com/phaseoteam/Phaseo/pull/2683)).

API & SDKs

* The Phaseo MCP app adds a model explorer with search, filters, comparisons of up to three models, cost estimates and code examples; see [CLI and MCP](https://phaseo.app/docs/v1/developers/cli-and-mcp.md) ([#2680](https://github.com/phaseoteam/Phaseo/pull/2680)).
* The Phaseo MCP app can run a text comparison after you review and approve its estimated cost ([#2680](https://github.com/phaseoteam/Phaseo/pull/2680)).

**October 1, 2026**

Models

![Clef model release artwork](https://mintcdn.com/aistats/EzqUOO_knTIHSpCt/images/model-releases/2026-10-01--cloudflare--clef.png?fit=max\&auto=format\&n=EzqUOO_knTIHSpCt\&q=85\&s=41a039f5ee61f54119e9a3583b473a5a)

* [Cloudflare: Clef](https://phaseo.app/models/cloudflare/clef)
* [Cloudflare: Clef Flash](https://phaseo.app/models/cloudflare/clef-flash)
* [Microsoft: MAI-Voice-2.1](https://phaseo.app/models/microsoft/mai-voice-2.1)
* [Microsoft: MAI-Voice-2.1-Flash](https://phaseo.app/models/microsoft/mai-voice-2.1-flash)
* [Microsoft: MAI-Transcribe-2-Streaming](https://phaseo.app/models/microsoft/mai-transcribe-2-streaming)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.