Code at the speed of thought – Zed is a high-performance, multiplayer code editor from the creators of Atom and Tree-sitter.
Update Zed-hosted model documentation (#60771)
morgankrey5f8a7413a317
diff --git a/docs/src/account/zed-hosted-models.md b/docs/src/account/zed-hosted-models.mdindex a74fd81b572..cc79a54ddf7 100644--- a/docs/src/account/zed-hosted-models.md+++ b/docs/src/account/zed-hosted-models.md@@ -1,16 +1,20 @@---title: Zed-Hosted Models - Zed-description: AI models available via Zed Pro including Claude, GPT-5.5, Gemini 3.1 Pro, and Gemini 3.5 Flash. Pricing, context windows, and tool call support.+description: AI models available via Zed Pro including Claude Fable 5, Claude Sonnet 5, GPT-5.6, and Gemini 3. Pricing, context windows, and tool call support.---# Zed-Hosted ModelsZed's plans offer hosted versions of major LLMs with higher rate limits than direct API access. Model availability is updated regularly. To use your own API keys instead, see [LLM Providers](../ai/llm-providers.md). For general setup, see [AI Quick Start](../ai/quick-start.md).-> **Note:** Claude Opus models, GPT-5.5 pro, and GPT-5.4 pro are only available on Zed Pro and Zed Business.+> **Note:** Claude Fable 5, Claude Opus models, GPT-5.5 pro, and GPT-5.4 pro are only available on Zed Pro and Zed Business.| Model | Provider | Token Type | Provider Price per 1M tokens | Zed Price per 1M tokens || ----------------- | --------- | ------------------- | ---------------------------- | ----------------------- |+| Claude Fable 5 | Anthropic | Input | $10.00 | $11.00 |+| | Anthropic | Output | $50.00 | $55.00 |+| | Anthropic | Input - Cache Write | $12.50 | $13.75 |+| | Anthropic | Input - Cache Read | $1.00 | $1.10 || Claude Opus 4.5 | Anthropic | Input | $5.00 | $5.50 || | Anthropic | Output | $25.00 | $27.50 || | Anthropic | Input - Cache Write | $6.25 | $6.875 |@@ -27,6 +31,10 @@ Zed's plans offer hosted versions of major LLMs with higher rate limits than dir| | Anthropic | Output | $25.00 | $27.50 || | Anthropic | Input - Cache Write | $6.25 | $6.875 || | Anthropic | Input - Cache Read | $0.50 | $0.55 |+| Claude Sonnet 5 | Anthropic | Input | $2.00 | $2.20 |+| | Anthropic | Output | $10.00 | $11.00 |+| | Anthropic | Input - Cache Write | $2.50 | $2.75 |+| | Anthropic | Input - Cache Read | $0.20 | $0.22 || Claude Sonnet 4.5 | Anthropic | Input | $3.00 | $3.30 || | Anthropic | Output | $15.00 | $16.50 || | Anthropic | Input - Cache Write | $3.75 | $4.125 |@@ -39,6 +47,18 @@ Zed's plans offer hosted versions of major LLMs with higher rate limits than dir| | Anthropic | Output | $5.00 | $5.50 || | Anthropic | Input - Cache Write | $1.25 | $1.375 || | Anthropic | Input - Cache Read | $0.10 | $0.11 |+| GPT-5.6 Sol | OpenAI | Input | $5.00 | $5.50 |+| | OpenAI | Output | $30.00 | $33.00 |+| | OpenAI | Input - Cache Write | $6.25 | $6.875 |+| | OpenAI | Cached Input | $0.50 | $0.55 |+| GPT-5.6 Terra | OpenAI | Input | $2.50 | $2.75 |+| | OpenAI | Output | $15.00 | $16.50 |+| | OpenAI | Input - Cache Write | $3.125 | $3.4375 |+| | OpenAI | Cached Input | $0.25 | $0.275 |+| GPT-5.6 Luna | OpenAI | Input | $1.00 | $1.10 |+| | OpenAI | Output | $6.00 | $6.60 |+| | OpenAI | Input - Cache Write | $1.25 | $1.375 |+| | OpenAI | Cached Input | $0.10 | $0.11 || GPT-5.5 pro | OpenAI | Input | $30.00 | $33.00 || | OpenAI | Output | $180.00 | $198.00 || GPT-5.5 | OpenAI | Input | $5.00 | $5.50 |@@ -71,17 +91,22 @@ Zed's plans offer hosted versions of major LLMs with higher rate limits than dir| Gemini 3 Flash | Google | Input | $0.50 | $0.55 || | Google | Output | $3.00 | $3.30 |-## Recent Model Retirements+> **Warn:** Anthropic retains prompts and outputs sent to Claude Fable 5 for at least 30 days for trust and safety purposes. The no-training commitment still applies, and Zed does not retain your prompts or outputs. See [Provider Safety Retention for Designated Models](../ai/privacy-and-security.md#provider-safety-retention).++The Claude Sonnet 5 prices shown above use Anthropic's introductory pricing through August 31, 2026.++## Recent Model Retirements {#recent-model-retirements}-As of February 19, 2026, Zed Pro serves newer model versions in place of the retired models below:+Zed no longer offers the models below:- Claude Opus 4.1 → Claude Opus 4.5, Claude Opus 4.6, Claude Opus 4.7, or Claude Opus 4.8- Claude Sonnet 4 → Claude Sonnet 4.5 or Claude Sonnet 4.6-- Claude Sonnet 3.7 (retired Feb 19) → Claude Sonnet 4.5 or Claude Sonnet 4.6+- Claude Sonnet 3.7 (retired February 19, 2026) → Claude Sonnet 4.5 or Claude Sonnet 4.6- GPT-5.1 and GPT-5 → GPT-5.2 or GPT-5.2-Codex- Gemini 2.5 Pro → Gemini 3.1 Pro-- Gemini 3 Pro → Gemini 3.1 Pro+- Gemini 3 Pro (retired March 26, 2026) → Gemini 3.1 Pro- Gemini 2.5 Flash → Gemini 3 Flash or Gemini 3.5 Flash+- Grok 4, Grok 4 Fast, Grok 4 Fast (Non-Reasoning), and Grok Code Fast 1 ([retired May 15, 2026](https://docs.x.ai/developers/migration/may-15-retirement)). Zed no longer offers hosted xAI models.## Usage {#usage}@@ -97,13 +122,18 @@ A context window is the maximum span of text and code an LLM can consider at onc| Model | Provider | Zed-Hosted Context Window || ----------------- | --------- | ------------------------- |+| Claude Fable 5 | Anthropic | 1M || Claude Opus 4.5 | Anthropic | 200k || Claude Opus 4.6 | Anthropic | 1M || Claude Opus 4.7 | Anthropic | 1M || Claude Opus 4.8 | Anthropic | 1M |+| Claude Sonnet 5 | Anthropic | 1M || Claude Sonnet 4.5 | Anthropic | 200k || Claude Sonnet 4.6 | Anthropic | 1M || Claude Haiku 4.5 | Anthropic | 200k |+| GPT-5.6 Sol | OpenAI | 272k input / 400k total |+| GPT-5.6 Terra | OpenAI | 272k input / 400k total |+| GPT-5.6 Luna | OpenAI | 272k input / 400k total || GPT-5.5 pro | OpenAI | 272k input / 400k total || GPT-5.5 | OpenAI | 272k input / 400k total || GPT-5.4 pro | OpenAI | 272k input / 400k total |diff --git a/docs/src/ai/ai-improvement.md b/docs/src/ai/ai-improvement.mdindex 72ff46b37fa..172a70e3f97 100644--- a/docs/src/ai/ai-improvement.md+++ b/docs/src/ai/ai-improvement.md@@ -10,7 +10,7 @@ Normal AI requests are not retained by Zed. Forprohibit training on your prompts or code context and require zero dataretention, except for[provider-designated models with safety retention](./privacy-and-security.md#provider-safety-retention),-such as Anthropic's Mythos-class models.+such as Anthropic's Covered Models.This page covers the cases where Zed may retain AI data because you explicitlyshared it or opted in.@@ -28,7 +28,7 @@ For the broader request path and provider data boundaries, seeZed-hosted model zero-data-retention and no-training commitments, including theexception for provider-designated models with safety retention (such as-Anthropic's Mythos-class models), are documented+Anthropic's Covered Models), are documentedon [AI Privacy](./privacy-and-security.md#data-retention-and-training).## Response Ratings and Feedback {#ai-feedback-with-ratings}diff --git a/docs/src/ai/privacy-and-security.md b/docs/src/ai/privacy-and-security.mdindex 354feba64af..3943206a37f 100644--- a/docs/src/ai/privacy-and-security.md+++ b/docs/src/ai/privacy-and-security.md@@ -15,24 +15,24 @@ Zed does not retain your prompts or code context by default. Forcommitments from model providers, and provider agreements require zero dataretention for inference requests except for[provider-designated models with safety retention](#provider-safety-retention),-such as Anthropic's Mythos-class models.+such as Anthropic's Covered Models.Zed only retains AI data when you explicitly share feedback or opt in totraining data collection.## AI Request Paths {#ai-request-paths}-| Path | Who handles model requests | What to know | Details |-| ------------------------------------------------------------ | ------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------- |-| [Zed-hosted models](../account/zed-hosted-models.md) | Zed routes requests to hosted model providers | Provider agreements prohibit training on your prompts or code context and require zero data retention for inference requests, except for provider-designated models with safety retention, such as Anthropic's Mythos-class models. | [Zed-hosted model commitments](#data-retention-and-training) |-| [Provider API keys](./use-api-access.md) | The configured provider | The provider handles requests under its own terms. Provider keys saved through Zed are stored in the system keychain, not in `settings.json`. | [Use API Access](./use-api-access.md) |-| [Existing subscriptions](./use-an-existing-subscription.md) | The subscription provider | The provider handles requests under the subscription terms. | [Use an Existing Subscription](./use-an-existing-subscription.md) |-| [Gateways](./use-a-gateway.md) | The configured gateway and upstream providers | The gateway and upstream providers handle requests under their own terms. | [Use a Gateway](./use-a-gateway.md) |-| [Local models](./use-a-local-model.md) | The local server or self-hosted endpoint | The local server handles requests according to how you configured that server. | [Use a Local Model](./use-a-local-model.md) |-| [External Agents](./external-agents.md) | The External Agent and its configured providers | The External Agent handles model requests under its own terms. Tool and MCP behavior depends on agent and ACP configuration. | [External Agents](./external-agents.md) |-| [Terminal Threads](./terminal-threads.md) | The CLI or TUI running in the terminal | The CLI or TUI owns its auth, model routing, tools, instructions, MCP configuration, and data handling. | [Terminal Threads](./terminal-threads.md) |-| [Edit Prediction](./edit-prediction.md) | The selected edit prediction provider | Each keystroke can send local editing context to the selected provider. Zeta requests are processed transiently unless training data collection is enabled; third-party providers follow their own terms. | [Edit Prediction](./edit-prediction.md), [Feedback and Training Data](./ai-improvement.md) |-| [Agent tools](./tools.md), [MCP](./mcp.md), and integrations | Zed, configured MCP servers, and external systems | Tools can read, edit, search, run commands, fetch URLs, or call external systems depending on profile, MCP server, and tool permission settings. | [Agent Profiles](./agent-profiles.md), [Tool Permissions](./tool-permissions.md), [MCP](./mcp.md) |-| Project trust and instructions | Zed and the trusted worktree | Project-local instructions and skills are loaded from trusted worktrees. External Agents and Terminal Threads may read their own instruction files. | [Worktree Trust](../worktree-trust.md), [Skills](./skills.md), [Instructions](./instructions.md) |+| Path | Who handles model requests | What to know | Details |+| ------------------------------------------------------------ | ------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------- |+| [Zed-hosted models](../account/zed-hosted-models.md) | Zed routes requests to hosted model providers | Provider agreements prohibit training on your prompts or code context and require zero data retention for inference requests, except for provider-designated models with safety retention, such as Anthropic's Covered Models. | [Zed-hosted model commitments](#data-retention-and-training) |+| [Provider API keys](./use-api-access.md) | The configured provider | The provider handles requests under its own terms. Provider keys saved through Zed are stored in the system keychain, not in `settings.json`. | [Use API Access](./use-api-access.md) |+| [Existing subscriptions](./use-an-existing-subscription.md) | The subscription provider | The provider handles requests under the subscription terms. | [Use an Existing Subscription](./use-an-existing-subscription.md) |+| [Gateways](./use-a-gateway.md) | The configured gateway and upstream providers | The gateway and upstream providers handle requests under their own terms. | [Use a Gateway](./use-a-gateway.md) |+| [Local models](./use-a-local-model.md) | The local server or self-hosted endpoint | The local server handles requests according to how you configured that server. | [Use a Local Model](./use-a-local-model.md) |+| [External Agents](./external-agents.md) | The External Agent and its configured providers | The External Agent handles model requests under its own terms. Tool and MCP behavior depends on agent and ACP configuration. | [External Agents](./external-agents.md) |+| [Terminal Threads](./terminal-threads.md) | The CLI or TUI running in the terminal | The CLI or TUI owns its auth, model routing, tools, instructions, MCP configuration, and data handling. | [Terminal Threads](./terminal-threads.md) |+| [Edit Prediction](./edit-prediction.md) | The selected edit prediction provider | Each keystroke can send local editing context to the selected provider. Zeta requests are processed transiently unless training data collection is enabled; third-party providers follow their own terms. | [Edit Prediction](./edit-prediction.md), [Feedback and Training Data](./ai-improvement.md) |+| [Agent tools](./tools.md), [MCP](./mcp.md), and integrations | Zed, configured MCP servers, and external systems | Tools can read, edit, search, run commands, fetch URLs, or call external systems depending on profile, MCP server, and tool permission settings. | [Agent Profiles](./agent-profiles.md), [Tool Permissions](./tool-permissions.md), [MCP](./mcp.md) |+| Project trust and instructions | Zed and the trusted worktree | Project-local instructions and skills are loaded from trusted worktrees. External Agents and Terminal Threads may read their own instruction files. | [Worktree Trust](../worktree-trust.md), [Skills](./skills.md), [Instructions](./instructions.md) |## Zed-Hosted Model Commitments {#data-retention-and-training}@@ -40,7 +40,7 @@ For Zed-hosted models, Zed has commitments from model providers that prohibittraining on your prompts or code context and require zero data retention forinference requests, except for[provider-designated models with safety retention](#provider-safety-retention),-such as Anthropic's Mythos-class models. The public provider documents linked below describe provider programs or default+such as Anthropic's Covered Models. The public provider documents linked below describe provider programs or defaultAPI terms; Zed-hosted model requests are governed by Zed's provider agreements.| Provider | No training reference | Zero-data-retention reference |@@ -53,11 +53,10 @@ API terms; Zed-hosted model requests are governed by Zed's provider agreements.Some providers require limited data retention for specific models as a conditionof offering them, on every platform where those models are available. Anthropic-retains prompts and outputs for models it designates as covered models (its-Mythos-class models, such as Claude Fable 5) for 30 days for trust and safety-purposes. Zed cannot opt out of this retention; it applies wherever these models-are served. See-[Anthropic's data retention practices for Mythos-class models](https://support.claude.com/en/articles/15425996-data-retention-practices-for-mythos-class-models).+retains prompts and outputs for models it designates as Covered Models, including+Claude Fable 5, for at least 30 days for trust and safety purposes. Zed cannot+opt out of this retention; it applies wherever these models are served. See+[Anthropic's data retention practices for Covered Models](https://support.claude.com/en/articles/15425996-data-retention-practices-for-covered-models).For these models:diff --git a/docs/src/business/privacy.md b/docs/src/business/privacy.mdindex 5b2e7e6e2e9..aeaaf3aa088 100644--- a/docs/src/business/privacy.md+++ b/docs/src/business/privacy.md@@ -36,7 +36,7 @@ Neither option is available to Zed Business members.## What data still leaves the organization-These controls cover what Zed stores and trains on. They don't change how AI inference works: when members use Zed's hosted models, prompts and code context are still sent to the relevant provider (Anthropic, OpenAI, Google, etc.) to generate responses. Zed maintains no-training commitments with these providers, and zero-data-retention commitments for all models except [provider-designated models with safety retention](../ai/privacy-and-security.md#provider-safety-retention), such as Anthropic's Mythos-class models. See [AI Privacy](../ai/privacy-and-security.md#data-retention-and-training) for details.+These controls cover what Zed stores and trains on. They don't change how AI inference works: when members use Zed's hosted models, prompts and code context are still sent to the relevant provider (Anthropic, OpenAI, Google, etc.) to generate responses. Zed maintains no-training commitments with these providers, and zero-data-retention commitments for all models except [provider-designated models with safety retention](../ai/privacy-and-security.md#provider-safety-retention), such as Anthropic's Covered Models. See [AI Privacy](../ai/privacy-and-security.md#data-retention-and-training) for details.[Bring-your-own-key](../ai/llm-providers.md), [gateways](../ai/use-a-gateway.md), [local or self-hosted models](../ai/use-a-local-model.md), [External Agents](../ai/external-agents.md), and [Terminal Threads](../ai/terminal-threads.md) are subject to each provider, gateway, server, agent, or CLI's own terms.