diff --git a/docs/src/SUMMARY.md b/docs/src/SUMMARY.md index abf6cc127254e5f3b7fdc91219b638a8bf50eec6..eb542497238780d93974cbc2627ce3466e23049b 100644 --- a/docs/src/SUMMARY.md +++ b/docs/src/SUMMARY.md @@ -58,9 +58,9 @@ - [LLM Providers](./ai/llm-providers.md) - [Agent Settings](./ai/agent-settings.md) - [Subscription](./ai/subscription.md) + - [Models](./ai/models.md) - [Plans and Usage](./ai/plans-and-usage.md) - [Billing](./ai/billing.md) - - [Models](./ai/models.md) - [Privacy and Security](./ai/privacy-and-security.md) - [AI Improvement](./ai/ai-improvement.md) diff --git a/docs/src/ai/agent-panel.md b/docs/src/ai/agent-panel.md index b6b748e2f58493cd62abbd3c6e7dc443182e992f..445b85337046d75b78ec4101051fc55c7dcf3752 100644 --- a/docs/src/ai/agent-panel.md +++ b/docs/src/ai/agent-panel.md @@ -10,7 +10,7 @@ If you're using the Agent Panel for the first time, you need to have at least on You can do that by: 1. [subscribing to our Pro plan](https://zed.dev/pricing), so you have access to our hosted models -2. [bringing your own API keys](./llm-providers.md#use-your-own-keys) for your desired provider +2. [using your own API keys](./llm-providers.md#use-your-own-keys), either from model providers like Anthropic or model gateways like OpenRouter. 3. using an external agent like [Gemini CLI](./external-agents.md#gemini-cli) or [Claude Code](./external-agents.md#claude-code) ## Overview {#overview} @@ -51,20 +51,18 @@ To view all historical conversations, reach for the `View All` option from withi ### Following the Agent {#following-the-agent} -Zed is built with collaboration natively integrated. -This approach extends to collaboration with AI as well. -To follow the agent reading through your codebase and performing edits, click on the "crosshair" icon button at the bottom left of the panel. +Zed is built with collaboration natively integrated, and this design pattern extends to collaboration with AI. To follow the agent as it reads and edits in your codebase, click on the "crosshair" icon button at the bottom left of the panel. You can also do that with the keyboard by pressing the `cmd`/`ctrl` modifier with `enter` when submitting a message. ### Get Notified {#get-notified} -If you send a prompt to the Agent and then move elsewhere, thus putting Zed in the background, you can be notified of whether its response is finished either via: +If you send a prompt to the Agent and then move elsewhere, putting Zed in the background, you can be notified when its response is finished via: - a visual notification that appears in the top right of your screen -- or a sound notification +- a sound notification -Both notification methods can be used together or individually according to your preference. +These notifications can be used together or individually, according to your preference. You can customize their behavior, including turning them off entirely, by using the `agent.notify_when_agent_waiting` and `agent.play_sound_when_agent_done` settings keys. @@ -76,8 +74,7 @@ To see which files specifically have been edited, expand the accordion bar that You're able to reject or accept each individual change hunk, or the whole set of changes made by the agent. -Edit diffs also appear in individual buffers. -So, if your active tab had edits made by the AI, you'll see diffs with the same accept/reject controls as in the multi-buffer. +Edit diffs also appear in individual buffers. If your active tab had edits made by the AI, you'll see diffs with the same accept/reject controls as in the multi-buffer. ## Adding Context {#adding-context} @@ -89,15 +86,16 @@ Pasting images as context is also supported by the Agent Panel. ### Token Usage {#token-usage} -Zed surfaces how many tokens you are consuming for your currently active thread nearby the profile selector in the panel's message editor. -Depending on how many pieces of context you add, your token consumption can grow rapidly. +Zed surfaces how many tokens you are consuming for your currently active thread nearby the profile selector in the panel's message editor. Depending on how many pieces of context you add, your token consumption can grow rapidly. -With that in mind, once you get close to the model's context window, a banner appears below the message editor suggesting to start a new thread with the current one summarized and added as context. +Once you approach the model's context window, a banner appears below the message editor suggesting to start a new thread with the current one summarized and added as context. You can also do this at any time with an ongoing thread via the "Agent Options" menu on the top right. ## Changing Models {#changing-models} -After you've configured your LLM providers—either via [a custom API key](./llm-providers.md#use-your-own-keys) or through [Zed's hosted models](./models.md)—you can switch between them by clicking on the model selector on the message editor or by using the {#kb agent::ToggleModelSelector} keybinding. +After you've configured your LLM providers—either via [a custom API key](./llm-providers.md) or through [Zed's hosted models](./models.md)—you can switch between them by clicking on the model selector on the message editor or by using the {#kb agent::ToggleModelSelector} keybinding. + +> The same model can be offered via multiple providers - for example, Claude Sonnet 4 is available via Zed Pro, OpenRouter, Anthropic directly, and more. Make sure you've selected the correct model **_provider_** for the model you'd like to use, delineated by the logo to the left of the model in the model selector. ## Using Tools {#using-tools} @@ -140,27 +138,19 @@ You can change that by setting this key to `true` in either your `settings.json` ### Model Support {#model-support} Tool calling needs to be individually supported by each model and model provider. -Therefore, despite the presence of tools, some models may not have the ability to pick them up yet in Zed. -You should see a "No tools" label if you select a model that falls into this case. - -We want to support all of them, though! -We may prioritize which ones to focus on based on popularity and user feedback, so feel free to help and contribute to fast-track those that don't fit this bill. +Therefore, despite the presence of tools, some models may not have the ability to pick them up yet in Zed. You should see a "No tools" label if you select a model that falls into this case. All [Zed's hosted models](./models.md) support tool calling out-of-the-box. ### MCP Servers {#mcp-servers} -Similarly to the built-in tools, some models may not support all tools included in a given MCP Server. -Zed's UI will inform about this via a warning icon that appears close to the model selector. +Similarly to the built-in tools, some models may not support all tools included in a given MCP Server. Zed's UI will inform about this via a warning icon that appears close to the model selector. ## Text Threads {#text-threads} -["Text Threads"](./text-threads.md) present your conversation with the LLM in a different format—as raw text. -With text threads, you have full control over the conversation data. -You can remove and edit responses from the LLM, swap roles, and include more context earlier in the conversation. +["Text Threads"](./text-threads.md) present your conversation with the LLM in a different format—as raw text. With text threads, you have full control over the conversation data. You can remove and edit responses from the LLM, swap roles, and include more context earlier in the conversation. -For users who have been with us for some time, you'll notice that text threads are our original assistant panel—users love it for the control it offers. -We do not plan to deprecate text threads, but it should be noted that if you want the AI to write to your code base autonomously, that's only available in the newer, and now default, "Threads". +For users who have been with us for some time, you'll notice that text threads are our original assistant panel—users love it for the control it offers. We do not plan to deprecate text threads, but it should be noted that if you want the AI to write to your code base autonomously, that's only available in the newer, and now default, "Threads". ## Errors and Debugging {#errors-and-debugging} @@ -170,15 +160,12 @@ You can also open threads as Markdown by clicking on the file icon button, to th ## Feedback {#feedback} -Every change we make to Zed's system prompt and tool set, needs to be backed by a thorough eval with good scores. - -Every time the LLM performs a weird change or investigates a certain topic in your code base incorrectly, it's an indication that there's an improvement opportunity. +Zed supports rating responses from the agent for feedback and improvement. > Note that rating responses will send your data related to that response to Zed's servers. > See [AI Improvement](./ai-improvement.md) and [Privacy and Security](./privacy-and-security.md) for more information about Zed's approach to AI improvement, privacy, and security. > **_If you don't want data persisted on Zed's servers, don't rate_**. We will not collect data for improving our Agentic offering without you explicitly rating responses. -The best way you can help influence the next change to Zed's system prompt and tools is by rating the LLM's response via the thumbs up/down buttons at the end of every response. -In case of a thumbs down, a new text area will show up where you can add more specifics about what happened. +The best way you can help influence the next change to Zed's system prompt and tools is by rating the LLM's response via the thumbs up/down buttons at the end of every response. In case of a thumbs down, a new text area will show up where you can add more specifics about what happened. You can provide feedback on the thread at any point after the agent responds, and multiple times within the same thread. diff --git a/docs/src/ai/agent-settings.md b/docs/src/ai/agent-settings.md index 4d12a87f95f6fd34a6dcfdd822b6e1f71dfeefa2..15f50c03c2fa972e55e00217e87f628093a9eb20 100644 --- a/docs/src/ai/agent-settings.md +++ b/docs/src/ai/agent-settings.md @@ -6,14 +6,13 @@ Learn about all the settings you can customize in Zed's Agent Panel. ### Default Model {#default-model} -If you're using [Zed's hosted LLM service](./plans-and-usage.md), it sets `claude-sonnet-4` as the default model. -But if you're not subscribed to it or simply just want to change it, you can do it so either via the model dropdown in the Agent Panel's bottom-right corner or by manually editing the `default_model` object in your settings: +If you're using [Zed's hosted LLM service](./subscription.md), it sets `claude-sonnet-4` as the default model for agentic work (agent panel, inline assistant) and `gpt-5-nano` as the default "fast" model (thread summarization, git commit messages). If you're not subscribed or want to change these defaults, you can manually edit the `default_model` object in your settings: ```json { "agent": { "default_model": { - "provider": "zed.dev", + "provider": "openai", "model": "gpt-4o" } } @@ -22,7 +21,7 @@ But if you're not subscribed to it or simply just want to change it, you can do ### Feature-specific Models {#feature-specific-models} -Assign distinct and specific models for the following AI-powered features in Zed: +You can assign distinct and specific models for the following AI-powered features: - Thread summary model: Used for generating thread summaries - Inline assistant model: Used for the inline assistant feature @@ -63,7 +62,7 @@ When configured, the inline assist UI will surface controls to cycle between the The models you specify here are always used in _addition_ to your [default model](#default-model). For example, the following configuration will generate two outputs for every assist. -One with Claude Sonnet 4 (the default model), and one with GPT-4o. +One with Claude Sonnet 4 (the default model), and one with GPT-5-mini. ```json { @@ -75,7 +74,7 @@ One with Claude Sonnet 4 (the default model), and one with GPT-4o. "inline_alternatives": [ { "provider": "zed.dev", - "model": "gpt-4o" + "model": "gpt-4-mini" } ] } diff --git a/docs/src/ai/ai-improvement.md b/docs/src/ai/ai-improvement.md index 5296b2a179ebe8ba8b7a966a54138cbde4650b4f..972b5908c08c6a7549553b0ae237714283c4b937 100644 --- a/docs/src/ai/ai-improvement.md +++ b/docs/src/ai/ai-improvement.md @@ -4,18 +4,30 @@ ### Opt-In -When using the Agent Panel, whether through Zed's hosted AI service or via connecting a non-Zed AI service via API key, Zed does not persistently store user content or use user content to evaluate and/or improve our AI features, unless it is explicitly shared with Zed. Each share is opt-in, and sharing once will not cause future content or data to be shared again. +When you use the Agent Panel through any of these means: + +- [Zed's hosted models](./subscription.md) +- [connecting a non-Zed AI service via API key](./llm-providers.md) +- using an [external agent](./external-agents.md) + +Zed does not persistently store user content or use user content to evaluate and/or improve our AI features, unless it is explicitly shared with Zed. Each share is opt-in, and sharing once will not cause future content or data to be shared again. > Note that rating responses will send your data related to that response to Zed's servers. > **_If you don't want data persisted on Zed's servers, don't rate_**. We will not collect data for improving our Agentic offering without you explicitly rating responses. -When using upstream services through Zed AI, we require assurances from our service providers that your user content won't be used for training models. For example, usage of Anthropic Claude 3.5 via Zed AI in the Assistant is governed by the [Anthropic Commercial Terms](https://www.anthropic.com/legal/commercial-terms) which includes the following: +When using upstream services through Zed's hosted models, we require assurances from our service providers that your user content won't be used for training models. + +| Provider | No Training Guarantee | Zero-Data Retention (ZDR) | +| --------- | ------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------- | +| Anthropic | [Yes](https://www.anthropic.com/legal/commercial-terms) | [Yes](https://privacy.anthropic.com/en/articles/8956058-i-have-a-zero-data-retention-agreement-with-anthropic-what-products-does-it-apply-to) | +| Google | [Yes](https://cloud.google.com/terms/service-terms) | **No**, in flight | +| OpenAI | [Yes](https://openai.com/enterprise-privacy/) | [Yes](https://platform.openai.com/docs/guides/your-data) | -> "Anthropic may not train models on Customer Content from paid Services." +> Zed's use of Gemini models is currently supported via [Google AI Studio](https://ai.google.dev/aistudio), which **_does not_** support ZDR. We're migrating to [Vertex AI](https://cloud.google.com/vertex-ai?hl=en), which **_does_**, and upon completion of that migration will offer ZDR to all users of Zed's hosted Google/Gemini models. -We also have a [zero-data retention agreement](https://privacy.anthropic.com/en/articles/8956058-i-have-a-zero-data-retention-agreement-with-anthropic-what-products-does-it-apply-to) with Anthropic. +> If ZDR from upstream model providers is important to you, _please do not use Gemini models at this time_. Your data will never be used for training purposes by any model providers hosted by Zed, however. -When you directly connect Zed with a non Zed AI service (e.g., via API key) Zed does not have control over how your data is used by that service provider. +When you use your own API keys or external agents, **Zed does not have control over how your data is used by that service provider.** You should reference your agreement with each service provider to understand what terms and conditions apply. ### Data we collect diff --git a/docs/src/ai/billing.md b/docs/src/ai/billing.md index d519b136aeea8c505979cde224d406ac995b65f0..64ff871ce1b629fad72d4ddd6f9c8f42f2bf92da 100644 --- a/docs/src/ai/billing.md +++ b/docs/src/ai/billing.md @@ -1,29 +1,42 @@ # Billing -We use Stripe as our billing and payments provider. All Pro plans require payment via credit card. +We use Stripe as our payments provider, and Orb for invoicing and metering. All Pro plans require payment via credit card or other supported payment method. For invoice-based billing, a Business plan is required. Contact [sales@zed.dev](mailto:sales@zed.dev) for more information. -## Settings {#settings} +## Billing Information {#settings} -You can access billing settings at [zed.dev/account](https://zed.dev/account). -Clicking the button under Account Settings will navigate you to Stripe’s secure portal, where you can update all billing-related settings and configuration. +You can access billing information and settings at [zed.dev/account](https://zed.dev/account). +Most of the page embeds information from our invoicing/metering partner, Orb (we're planning on a more native experience soon!). ## Billing Cycles {#billing-cycles} -Zed is billed on a monthly basis based on the date you initially subscribe. +Zed is billed on a monthly basis based on the date you initially subscribe. You'll receive _at least_ one invoice from Zed each month you're subscribed to Zed Pro, and more than one if you use more than $10 in incremental token spend within the month. -We’ll also bill in-month for additional prompts used beyond your plan’s prompt limit, if usage exceeds $20 before month end. -See [usage-based pricing](./plans-and-usage.md#ubp) for more. +## Threshold Billing {#threshold-billing} + +Zed utilizes threshold billing to ensure timely collection of owed monies and prevent abuse. Every time your usage of Zed's hosted models crosses a $10 spend threshold, a new invoice is generated, and the threshold resets to $0. + +For example, + +- You subscribe on February 1. Your first invoice is $10. +- You use $12 of incremental tokens in the month of February, with the first $10 spent on February 15. You'll receive an invoice for $10 on February 15 +- On March 1, you receive an invoice for $12: $10 (March Pro subscription) and $2 in leftover token spend, since your usage didn't cross the $10 threshold. + +## Payment Failures {#payment-failures} + +If payment of an invoice fails, Zed will block usage of our hosted models until the payment is complete. Email [billing-support@zed.dev](mailto:billing-support@zed.dev) for assistance. ## Invoice History {#invoice-history} -You can access your invoice history by navigating to [zed.dev/account](https://zed.dev/account) and clicking "Manage" on your subscription. +You can access your invoice history by navigating to [zed.dev/account](https://zed.dev/account) and clicking `Invoice history` within the embedded Orb portal. -From Stripe’s secure portal, you can download all current and historical invoices. +If you require historical Stripe invoices, email [billing-support@zed.dev](mailto:billing-support@zed.dev) ## Updating Billing Information {#updating-billing-info} -You can update your payment method, company name, address, and tax information through the billing portal. +Email [billing-support@zed.dev](mailto:billing-support@zed.dev) for help updating payment methods, names, addresses, and tax information. + +> We'll be updating our account page shortly to allow for self-service updates. Stay tuned! Please note that changes to billing information will **only** affect future invoices — **we cannot modify historical invoices**. @@ -31,7 +44,7 @@ Please note that changes to billing information will **only** affect future invo Zed partners with [Sphere](https://www.getsphere.com/) to calculate indirect tax rate for invoices, based on customer location and the product being sold. Tax is listed as a separate line item on invoices, based preferentially on your billing address, followed by the card issue country known to Stripe. -If you have a VAT/GST ID, you can add it at [zed.dev/account](https://zed.dev/account) by clicking "Manage" on your subscription. Check the box that denotes you as a business. +If you have a VAT/GST ID, you can add it at during checkout. Check the box that denotes you as a business. Please note that changes to VAT/GST IDs and address will **only** affect future invoices — **we cannot modify historical invoices**. Questions or issues can be directed to [billing-support@zed.dev](mailto:billing-support@zed.dev). diff --git a/docs/src/ai/configuration.md b/docs/src/ai/configuration.md index d28a7e8ed006b1c788cc0f649362bae41879a99b..c11a0fd65c45ce46598596182fbf8fb0c147380a 100644 --- a/docs/src/ai/configuration.md +++ b/docs/src/ai/configuration.md @@ -1,8 +1,11 @@ # Configuration -When using AI in Zed, you can customize several aspects: +When using AI in Zed, you can configure multiple dimensions: -1. Which [LLM providers](./llm-providers.md) you can use +1. Which LLM providers you can use + - Zed's hosted models, which require [authentication](../accounts.md) and [subscription](./subscription.md) + - [Using your own API keys](./llm-providers.md), which do not + - Using [external agents like Claude Code](./external-agents.md), which do not 2. [Model parameters and usage](./agent-settings.md#model-settings) 3. [Interactions with the Agent Panel](./agent-settings.md#agent-panel-settings) diff --git a/docs/src/ai/external-agents.md b/docs/src/ai/external-agents.md index bc7768c6081ad7a32eb1fd780750a48c4b9200f0..abe14865902ee261b157cf653b0d556cf83d7c71 100644 --- a/docs/src/ai/external-agents.md +++ b/docs/src/ai/external-agents.md @@ -5,6 +5,8 @@ Zed supports terminal-based agents through the [Agent Client Protocol (ACP)](htt Currently, [Gemini CLI](https://github.com/google-gemini/gemini-cli) serves as the reference implementation. [Claude Code](https://www.anthropic.com/claude-code) is also included by default, and you can [add custom ACP-compatible agents](#add-custom-agents) as well. +Zed's affordance for external agents is strictly UI-based; the billing and legal/terms arrangement is directly between you and the agent provider. Zed does not charge for use of external agents, and our [zero-data retention agreements/privacy guarantees](./ai-improvement.md) are **_only_** applicable for Zed's hosted models. + ## Gemini CLI {#gemini-cli} Zed provides the ability to run [Gemini CLI](https://github.com/google-gemini/gemini-cli) directly in the [agent panel](./agent-panel.md). diff --git a/docs/src/ai/inline-assistant.md b/docs/src/ai/inline-assistant.md index da894e2cd87faf6ce8afa9c54a5f2d55bcd07827..41923e85da09c2eed067d40518c89088d653b7b7 100644 --- a/docs/src/ai/inline-assistant.md +++ b/docs/src/ai/inline-assistant.md @@ -14,8 +14,6 @@ Give the Inline Assistant context the same way you can in [the Agent Panel](./ag A useful pattern here is to create a thread in the Agent Panel, and then mention that thread with `@thread` in the Inline Assistant to include it as context. -> The Inline Assistant is limited to normal mode context windows ([see Models](./models.md) for more). - ## Prefilling Prompts To create a custom keybinding that prefills a prompt, you can add the following format in your keymap: diff --git a/docs/src/ai/llm-providers.md b/docs/src/ai/llm-providers.md index 09f67cc9c123a968705a834f9d1c5a2e855a782f..aeed1be17370c28ad67d8ffb7d49fadc5a77cdce 100644 --- a/docs/src/ai/llm-providers.md +++ b/docs/src/ai/llm-providers.md @@ -6,7 +6,7 @@ You can do that by either subscribing to [one of Zed's plans](./plans-and-usage. ## Use Your Own Keys {#use-your-own-keys} -If you already have an API key for an existing LLM provider—say Anthropic or OpenAI, for example—you can insert them into Zed and use the full power of the Agent Panel **_for free_**. +If you already have an API key for an existing LLM provider, like Anthropic or OpenAI, you can add them to Zed and use the full power of the Agent Panel **_for free_**. To add an existing API key to a given provider, go to the Agent Panel settings (`agent: open settings`), look for the desired provider, paste the key into the input, and hit enter. @@ -14,7 +14,7 @@ To add an existing API key to a given provider, go to the Agent Panel settings ( ## Supported Providers -Here's all the supported LLM providers for which you can use your own API keys: +Zed offers an extensive list of "use your own key" LLM providers - [Amazon Bedrock](#amazon-bedrock) - [Anthropic](#anthropic) diff --git a/docs/src/ai/models.md b/docs/src/ai/models.md index 8d46d0b8d161ae23c77322d13244660142eab56a..8bebf15cb107298c00aa9b6fe405500545994be3 100644 --- a/docs/src/ai/models.md +++ b/docs/src/ai/models.md @@ -1,51 +1,58 @@ # Models -Zed’s plans offer hosted versions of major LLM’s, generally with higher rate limits than individual API keys. +Zed’s plans offer hosted versions of major LLMs, generally with higher rate limits than using your API keys. We’re working hard to expand the models supported by Zed’s subscription offerings, so please check back often. -| Model | Provider | Burn Mode | Context Window | Price per Prompt | Price per Request | -| ----------------- | --------- | --------- | -------------- | ---------------- | ----------------- | -| Claude 3.5 Sonnet | Anthropic | ❌ | 60k | $0.04 | N/A | -| Claude 3.7 Sonnet | Anthropic | ❌ | 120k | $0.04 | N/A | -| Claude 3.7 Sonnet | Anthropic | ✅ | 200k | N/A | $0.05 | -| Claude Sonnet 4 | Anthropic | ❌ | 120k | $0.04 | N/A | -| Claude Sonnet 4 | Anthropic | ✅ | 200k | N/A | $0.05 | -| Claude Opus 4 | Anthropic | ❌ | 120k | $0.20 | N/A | -| Claude Opus 4 | Anthropic | ✅ | 200k | N/A | $0.25 | -| Claude Opus 4.1 | Anthropic | ❌ | 120k | $0.20 | N/A | -| Claude Opus 4.1 | Anthropic | ✅ | 200k | N/A | $0.25 | - -> Note: Because of the 5x token cost for [Opus relative to Sonnet](https://www.anthropic.com/pricing#api), each Opus 4 and 4.1 prompt consumes 5 prompts against your billing meter +| Model | Provider | Token Type | Provider Price per 1M tokens | Zed Price per 1M tokens | +| ----------------- | --------- | ------------------- | ---------------------------- | ----------------------- | +| Claude Opus 4.1 | Anthropic | Input | $15.00 | $16.50 | +| | Anthropic | Output | $75.00 | $82.50 | +| | Anthropic | Input - Cache Write | $18.75 | $20.625 | +| | Anthropic | Input - Cache Read | $1.50 | $1.65 | +| Claude Sonnet 4 | Anthropic | Input | $3.00 | $3.30 | +| | Anthropic | Output | $15.00 | $16.50 | +| | Anthropic | Input - Cache Write | $3.75 | $4.125 | +| | Anthropic | Input - Cache Read | $0.30 | $0.33 | +| Claude Sonnet 3.7 | Anthropic | Input | $3.00 | $3.30 | +| | Anthropic | Output | $15.00 | $16.50 | +| | Anthropic | Input - Cache Write | $3.75 | $4.125 | +| | Anthropic | Input - Cache Read | $0.30 | $0.33 | +| GPT-5 | OpenAI | Input | $1.25 | $1.375 | +| | OpenAI | Output | $10.00 | $11.00 | +| | OpenAI | Cached Input | $0.125 | $0.1375 | +| GPT-5 mini | OpenAI | Input | $0.25 | $0.275 | +| | OpenAI | Output | $2.00 | $2.20 | +| | OpenAI | Cached Input | $0.025 | $0.0275 | +| GPT-5 nano | OpenAI | Input | $0.05 | $0.055 | +| | OpenAI | Output | $0.40 | $0.44 | +| | OpenAI | Cached Input | $0.005 | $0.0055 | +| Gemini 2.5 Pro | Google | Input | $1.25 | $1.375 | +| | Google | Output | $10.00 | $11.00 | +| Gemini 2.5 Flash | Google | Input | $0.30 | $0.33 | +| | Google | Output | $2.50 | $2.75 | ## Usage {#usage} -The models above can be used with the prompts included in your plan. For models not marked with [“Burn Mode”](#burn-mode), each prompt is counted against the monthly limit of your plan. - -If you’ve exceeded your limit for the month, and are on a paid plan, you can enable usage-based pricing to continue using models for the rest of the month. See [Plans and Usage](./plans-and-usage.md) for more information. - -Non-Burn Mode usage will use up to 25 tool calls per one prompt. If your prompt extends beyond 25 tool calls, Zed will ask if you’d like to continue, which will consume a second prompt. - -## Burn Mode {#burn-mode} - -> Note: "Burn Mode" is the new name for what was previously called "Max Mode". - -In Burn Mode, we enable models to use [large context windows](#context-windows), unlimited tool calls, and other capabilities for expanded reasoning, to allow an unfettered agentic experience. +Any usage of a Zed-hosted model will be billed at the Zed Price (rightmost column above). See [Plans and Usage](./plans-and-usage.md) for details on Zed's plans and limits for use of hosted models. -Because of the increased cost to Zed, each subsequent request beyond the initial user prompt in Burn Mode models is counted as a prompt for metering. - -In addition, usage-based pricing per request is slightly more expensive for Burn Mode models than usage-based pricing per prompt for regular models. - -> Note that the Agent Panel using a Burn Mode model may consume a good bit of your monthly prompt capacity, if many tool calls are used. -> We encourage you to think through what model is best for your needs before leaving the Agent Panel to work. - -By default, all threads and [text threads](./text-threads.md) start in normal mode. -However, you can use the `agent.preferred_completion_mode` setting to have Burn Mode activated by default. +> We encourage you to think through what model is best for your needs before leaving the Agent Panel to work. All LLMs can "spiral" and occasionally enter unending loops that require user intervention. ## Context Windows {#context-windows} A context window is the maximum span of text and code an LLM can consider at once, including both the input prompt and output generated by the model. -In [Burn Mode](#burn-mode), we increase context window size to allow models to have enhanced reasoning capabilities. +| Model | Provider | Zed-Hosted Context Window | +| ----------------- | --------- | ------------------------- | +| Claude Opus 4.1 | Anthropic | 200k | +| Claude Sonnet 4 | Anthropic | 200k | +| Claude Sonnet 3.7 | Anthropic | 200k | +| GPT-5 | OpenAI | 400k | +| GPT-5 mini | OpenAI | 400k | +| GPT-5 nano | OpenAI | 400k | +| Gemini 2.5 Pro | Google | 200k | +| Gemini 2.5 Flash | Google | 200k | + +> We're planning on expanding supported context windows for hosted Sonnet 4 and Gemini 2.5 Pro/Flash in the near future. Stay tuned! Each Agent thread and text thread in Zed maintains its own context window. The more prompts, attached files, and responses included in a session, the larger the context window grows. @@ -55,7 +62,3 @@ For best results, it’s recommended you take a purpose-based approach to Agent ## Tool Calls {#tool-calls} Models can use [tools](./tools.md) to interface with your code, search the web, and perform other useful functions. - -In [Burn Mode](#burn-mode), models can use an unlimited number of tools per prompt, with each tool call counting as a prompt for metering purposes. - -For non-Burn Mode models, you'll need to interact with the model every 25 tool calls to continue, at which point a new prompt will be counted against your plan limit. diff --git a/docs/src/ai/overview.md b/docs/src/ai/overview.md index 55d37ea3526173b6bf88adc0f15754be51bf6866..ca06a4b1ed53d1fc87136a1d5e82da35552082aa 100644 --- a/docs/src/ai/overview.md +++ b/docs/src/ai/overview.md @@ -8,7 +8,7 @@ Learn how to get started using AI with Zed and all its capabilities. - [External Agents](./external-agents.md): Learn how to plug in your favorite agent into Zed. -- [Subscription](./subscription.md): Learn about Zed's hosted model service and other billing-related information. +- [Subscription](./subscription.md): Learn about Zed's hosted models and other billing-related information. - [Privacy and Security](./privacy-and-security.md): Understand how Zed handles privacy and security with AI features. diff --git a/docs/src/ai/plans-and-usage.md b/docs/src/ai/plans-and-usage.md index 1e6616c79b80489b91e4f92c13b9c5fe39ff1af5..450ece0ca78855b4be4641ceb0b29322d98880c6 100644 --- a/docs/src/ai/plans-and-usage.md +++ b/docs/src/ai/plans-and-usage.md @@ -1,8 +1,5 @@ # Plans and Usage -To view your current usage, you can visit your account at [zed.dev/account](https://zed.dev/account). -You’ll also find usage meters in-product when you’re nearing the limit for your plan or trial. - ## Available Plans {#plans} For costs and more information on pricing, visit [Zed’s pricing page](https://zed.dev/pricing). @@ -11,29 +8,20 @@ Please note that if you’re interested in just using Zed as the world’s faste ## Usage {#usage} -- A `prompt` in Zed is an input from the user, initiated by pressing enter, composed of one or many `requests`. A `prompt` can be initiated from the Agent Panel, or via Inline Assist. -- A `request` in Zed is a response to a `prompt`, plus any tool calls that are initiated as part of that response. There may be one `request` per `prompt`, or many. - -Most models offered by Zed are metered per-prompt. -Some models that use large context windows and unlimited tool calls ([“Burn Mode”](./models.md#burn-mode)) count each individual request within a prompt against your prompt limit, since the agentic work spawned by the prompt is expensive to support. - -See [the Models page](./models.md) for a list of which subset of models are metered by request. +Usage of Zed's hosted models is measured on a token basis, converted to dollars at the rates lists on [the Models page](./models.md) (list price from the provider, +10%). -Plans come with a set amount of prompts included, with the number varying depending on the plan you’ve selected. +Zed Pro comes with $5 of monthly dollar credit. A trial of Zed Pro includes $20 of credit, usable for 14 days. Monthly included credit resets on your monthly billing date. -## Usage-Based Pricing {#ubp} +To view your current usage, you can visit your account at [zed.dev/account](https://zed.dev/account). Information from our metering and billing provider, Orb, is embedded on that page. -You may opt in to usage-based pricing for prompts that exceed what is included in your paid plan from [your account page](https://zed.dev/account). +## Spend Limits {#usage-spend-limits} -Usage-based pricing is only available with a paid plan, and is exclusively opt-in. -From the dashboard, you can toggle usage-based pricing for usage exceeding your paid plan. -You can also configure a spend limit in USD. -Once the spend limit is hit, we’ll stop any further usage until your prompt limit resets. +At the top of [the Account page](https://zed.dev/account), you'll find an input for `Maximum Token Spend`. The dollar amount here specifies your _monthly_ limit for spend on tokens, _not counting_ the $5/month included with your Pro subscription. -We will bill for additional prompts when you’ve made prompts totaling $20, or when your billing date occurs, whichever comes first. +The default value for all Pro users is $10, for a total monthly spend with Zed of $20 ($10 for your Pro subscription, $10 in incremental token spend). This can be set to $0 to limit your spend with Zed to exactly $10/month. If you adjust this limit _higher_ than $10 and consume more than $10 of incremental token spend, you'll be billed via [threshold billing](./billing.md#threshold-billing). -Cost per request for each model can be found on [the models page](./models.md). +Once the spend limit is hit, we’ll stop any further usage until your token spend limit resets. ## Business Usage {#business-usage} -Email [sales@zed.dev](mailto:sales@zed.dev) with any questions on business plans, metering, and usage-based pricing. +Email [sales@zed.dev](mailto:sales@zed.dev) with any questions on business plans. diff --git a/docs/src/ai/privacy-and-security.md b/docs/src/ai/privacy-and-security.md index 5e105a1e5618acc87a0be04290fb7972b6bd017e..23166df1d7e2e188e68d7ecafa98ac92f8e704a4 100644 --- a/docs/src/ai/privacy-and-security.md +++ b/docs/src/ai/privacy-and-security.md @@ -8,7 +8,7 @@ We believe in opt-in data sharing as the default in building AI products, rather As an open-source product, we believe in maximal transparency, and invite you to examine our codebase. If you find issues, we encourage you to share them with us. -It is entirely possible to use Zed without sharing any data with us and without authenticating into the product. We're happy to always support this desired use pattern. +It is entirely possible to use Zed, including Zed's AI capabilities, without sharing any data with us and without authenticating into the product. We're happy to always support this desired use pattern. ## Documentation diff --git a/docs/src/ai/subscription.md b/docs/src/ai/subscription.md index 078fe43384a4fc51b0413ef0bfa8fc7a8ddb1e38..704fdc0ce3a42e92f5a7b5cb58b2bfa1b4fbfb92 100644 --- a/docs/src/ai/subscription.md +++ b/docs/src/ai/subscription.md @@ -2,12 +2,12 @@ Zed's hosted models are offered via subscription to Zed Pro or Zed Business. -> Using your own API keys is _free_—you do not need to subscribe to a Zed plan to use our AI features with your own keys. +> Using [your own API keys](./llm-providers.md), and [external agents](./external-agents.md), is _free_ — you do not need to subscribe to a Zed plan to use AI features. See the following pages for specific aspects of our subscription offering: -- [Plans and Usage](./plans-and-usage.md): Outlines Zed's plans, how usage is measured, and usage-based pricing for overages. +- [Models](./models.md): Overview of the models offered by Zed's subscriptions. -- [Billing](./billing.md): Billing policies and procedures, and how to update or view various billing settings. +- [Plans and Usage](./plans-and-usage.md): Outlines Zed's plans and how usage is measured. -- [Models](./models.md): Overview of the models offered by Zed's subscriptions. +- [Billing](./billing.md): Billing policies and procedures, and how to update or view various billing settings. diff --git a/docs/src/git.md b/docs/src/git.md index cccbad9b2e37ba55dc45f1f100883437759727f0..f40040bec83226b19c17d9efdaf9241032dca7a5 100644 --- a/docs/src/git.md +++ b/docs/src/git.md @@ -79,7 +79,7 @@ In there, you can use the "Uncommit" button, which performs the `git reset HEAD Zed currently supports LLM-powered commit message generation. You can ask AI to generate a commit message by focusing on the message editor within the Git Panel and either clicking on the pencil icon in the bottom left, or reaching for the {#action git::GenerateCommitMessage} ({#kb git::GenerateCommitMessage}) keybinding. -> Note that you need to have an LLM provider configured. Visit [the AI configuration page](./ai/configuration.md) to learn how to do so. +> Note that you need to have an LLM provider configured for billing purposes, either via your own API keys or trialing/paying for Zed's hosted AI models. Visit [the AI configuration page](./ai/configuration.md) to learn how to do so. You can specify your preferred model to use by providing a `commit_message_model` agent setting. See [Feature-specific models](./ai/agent-settings.md#feature-specific-models) for more information.