1use anyhow::{Context as _, Result, anyhow};
2use collections::BTreeMap;
3use credentials_provider::CredentialsProvider;
4use futures::{FutureExt, Stream, StreamExt, future, future::BoxFuture};
5use google_ai::{
6 FunctionDeclaration, GenerateContentResponse, GoogleModelMode, Part, SystemInstruction,
7 ThinkingConfig, UsageMetadata,
8};
9use gpui::{AnyView, App, AsyncApp, Context, Entity, SharedString, Task, Window};
10use http_client::HttpClient;
11use language_model::{
12 AuthenticateError, ConfigurationViewTargetAgent, LanguageModelCompletionError,
13 LanguageModelCompletionEvent, LanguageModelToolChoice, LanguageModelToolSchemaFormat,
14 LanguageModelToolUse, LanguageModelToolUseId, MessageContent, StopReason,
15};
16use language_model::{
17 LanguageModel, LanguageModelId, LanguageModelName, LanguageModelProvider,
18 LanguageModelProviderId, LanguageModelProviderName, LanguageModelProviderState,
19 LanguageModelRequest, RateLimiter, Role,
20};
21use schemars::JsonSchema;
22use serde::{Deserialize, Serialize};
23pub use settings::GoogleAvailableModel as AvailableModel;
24use settings::{Settings, SettingsStore};
25use std::pin::Pin;
26use std::sync::{
27 Arc, LazyLock,
28 atomic::{self, AtomicU64},
29};
30use strum::IntoEnumIterator;
31use ui::{Icon, IconName, List, Tooltip, prelude::*};
32use ui_input::SingleLineInput;
33use util::{ResultExt, truncate_and_trailoff};
34use zed_env_vars::EnvVar;
35
36use crate::api_key::ApiKey;
37use crate::api_key::ApiKeyState;
38use crate::ui::InstructionListItem;
39
40const PROVIDER_ID: LanguageModelProviderId = language_model::GOOGLE_PROVIDER_ID;
41const PROVIDER_NAME: LanguageModelProviderName = language_model::GOOGLE_PROVIDER_NAME;
42
43#[derive(Default, Clone, Debug, PartialEq)]
44pub struct GoogleSettings {
45 pub api_url: String,
46 pub available_models: Vec<AvailableModel>,
47}
48
49#[derive(Clone, Copy, Debug, Default, PartialEq, Serialize, Deserialize, JsonSchema)]
50#[serde(tag = "type", rename_all = "lowercase")]
51pub enum ModelMode {
52 #[default]
53 Default,
54 Thinking {
55 /// The maximum number of tokens to use for reasoning. Must be lower than the model's `max_output_tokens`.
56 budget_tokens: Option<u32>,
57 },
58}
59
60pub struct GoogleLanguageModelProvider {
61 http_client: Arc<dyn HttpClient>,
62 state: Entity<State>,
63}
64
65pub struct State {
66 api_key_state: ApiKeyState,
67}
68
69const GEMINI_API_KEY_VAR_NAME: &str = "GEMINI_API_KEY";
70const GOOGLE_AI_API_KEY_VAR_NAME: &str = "GOOGLE_AI_API_KEY";
71
72static API_KEY_ENV_VAR: LazyLock<EnvVar> = LazyLock::new(|| {
73 // Try GEMINI_API_KEY first as primary, fallback to GOOGLE_AI_API_KEY
74 EnvVar::new(GEMINI_API_KEY_VAR_NAME.into()).or(EnvVar::new(GOOGLE_AI_API_KEY_VAR_NAME.into()))
75});
76
77impl State {
78 const fn is_authenticated(&self) -> bool {
79 self.api_key_state.has_key()
80 }
81
82 fn set_api_key(&mut self, api_key: Option<String>, cx: &mut Context<Self>) -> Task<Result<()>> {
83 let api_url = GoogleLanguageModelProvider::api_url(cx);
84 self.api_key_state
85 .store(api_url, api_key, |this| &mut this.api_key_state, cx)
86 }
87
88 fn authenticate(&mut self, cx: &mut Context<Self>) -> Task<Result<(), AuthenticateError>> {
89 let api_url = GoogleLanguageModelProvider::api_url(cx);
90 self.api_key_state.load_if_needed(
91 api_url,
92 &API_KEY_ENV_VAR,
93 |this| &mut this.api_key_state,
94 cx,
95 )
96 }
97}
98
99impl GoogleLanguageModelProvider {
100 pub fn new(http_client: Arc<dyn HttpClient>, cx: &mut App) -> Self {
101 let state = cx.new(|cx| {
102 cx.observe_global::<SettingsStore>(|this: &mut State, cx| {
103 let api_url = Self::api_url(cx);
104 this.api_key_state.handle_url_change(
105 api_url,
106 &API_KEY_ENV_VAR,
107 |this| &mut this.api_key_state,
108 cx,
109 );
110 cx.notify();
111 })
112 .detach();
113 State {
114 api_key_state: ApiKeyState::new(Self::api_url(cx)),
115 }
116 });
117
118 Self { http_client, state }
119 }
120
121 fn create_language_model(&self, model: google_ai::Model) -> Arc<dyn LanguageModel> {
122 Arc::new(GoogleLanguageModel {
123 id: LanguageModelId::from(model.id().to_string()),
124 model,
125 state: self.state.clone(),
126 http_client: self.http_client.clone(),
127 request_limiter: RateLimiter::new(4),
128 })
129 }
130
131 pub fn api_key_for_gemini_cli(cx: &mut App) -> Task<Result<String>> {
132 if let Some(key) = API_KEY_ENV_VAR.value.clone() {
133 return Task::ready(Ok(key));
134 }
135 let credentials_provider = <dyn CredentialsProvider>::global(cx);
136 let api_url = Self::api_url(cx).to_string();
137 cx.spawn(async move |cx| {
138 Ok(
139 ApiKey::load_from_system_keychain(&api_url, credentials_provider.as_ref(), cx)
140 .await?
141 .key()
142 .to_string(),
143 )
144 })
145 }
146
147 fn settings(cx: &App) -> &GoogleSettings {
148 &crate::AllLanguageModelSettings::get_global(cx).google
149 }
150
151 fn api_url(cx: &App) -> SharedString {
152 let api_url = &Self::settings(cx).api_url;
153 if api_url.is_empty() {
154 google_ai::API_URL.into()
155 } else {
156 SharedString::new(api_url.as_str())
157 }
158 }
159}
160
161impl LanguageModelProviderState for GoogleLanguageModelProvider {
162 type ObservableEntity = State;
163
164 fn observable_entity(&self) -> Option<Entity<Self::ObservableEntity>> {
165 Some(self.state.clone())
166 }
167}
168
169impl LanguageModelProvider for GoogleLanguageModelProvider {
170 fn id(&self) -> LanguageModelProviderId {
171 PROVIDER_ID
172 }
173
174 fn name(&self) -> LanguageModelProviderName {
175 PROVIDER_NAME
176 }
177
178 fn icon(&self) -> IconName {
179 IconName::AiGoogle
180 }
181
182 fn default_model(&self, _cx: &App) -> Option<Arc<dyn LanguageModel>> {
183 Some(self.create_language_model(google_ai::Model::default()))
184 }
185
186 fn default_fast_model(&self, _cx: &App) -> Option<Arc<dyn LanguageModel>> {
187 Some(self.create_language_model(google_ai::Model::default_fast()))
188 }
189
190 fn provided_models(&self, cx: &App) -> Vec<Arc<dyn LanguageModel>> {
191 let mut models = BTreeMap::default();
192
193 // Add base models from google_ai::Model::iter()
194 for model in google_ai::Model::iter() {
195 if !matches!(model, google_ai::Model::Custom { .. }) {
196 models.insert(model.id().to_string(), model);
197 }
198 }
199
200 // Override with available models from settings
201 for model in &GoogleLanguageModelProvider::settings(cx).available_models {
202 models.insert(
203 model.name.clone(),
204 google_ai::Model::Custom {
205 name: model.name.clone(),
206 display_name: model.display_name.clone(),
207 max_tokens: model.max_tokens,
208 mode: model.mode.unwrap_or_default(),
209 },
210 );
211 }
212
213 models
214 .into_values()
215 .map(|model| {
216 Arc::new(GoogleLanguageModel {
217 id: LanguageModelId::from(model.id().to_string()),
218 model,
219 state: self.state.clone(),
220 http_client: self.http_client.clone(),
221 request_limiter: RateLimiter::new(4),
222 }) as Arc<dyn LanguageModel>
223 })
224 .collect()
225 }
226
227 fn is_authenticated(&self, cx: &App) -> bool {
228 self.state.read(cx).is_authenticated()
229 }
230
231 fn authenticate(&self, cx: &mut App) -> Task<Result<(), AuthenticateError>> {
232 self.state.update(cx, |state, cx| state.authenticate(cx))
233 }
234
235 fn configuration_view(
236 &self,
237 target_agent: language_model::ConfigurationViewTargetAgent,
238 window: &mut Window,
239 cx: &mut App,
240 ) -> AnyView {
241 cx.new(|cx| ConfigurationView::new(self.state.clone(), target_agent, window, cx))
242 .into()
243 }
244
245 fn reset_credentials(&self, cx: &mut App) -> Task<Result<()>> {
246 self.state
247 .update(cx, |state, cx| state.set_api_key(None, cx))
248 }
249}
250
251pub struct GoogleLanguageModel {
252 id: LanguageModelId,
253 model: google_ai::Model,
254 state: Entity<State>,
255 http_client: Arc<dyn HttpClient>,
256 request_limiter: RateLimiter,
257}
258
259impl GoogleLanguageModel {
260 fn stream_completion(
261 &self,
262 request: google_ai::GenerateContentRequest,
263 cx: &AsyncApp,
264 ) -> BoxFuture<
265 'static,
266 Result<futures::stream::BoxStream<'static, Result<GenerateContentResponse>>>,
267 > {
268 let http_client = self.http_client.clone();
269
270 let Ok((api_key, api_url)) = self.state.read_with(cx, |state, cx| {
271 let api_url = GoogleLanguageModelProvider::api_url(cx);
272 (state.api_key_state.key(&api_url), api_url)
273 }) else {
274 return future::ready(Err(anyhow!("App state dropped"))).boxed();
275 };
276
277 async move {
278 let api_key = api_key.context("Missing Google API key")?;
279 let request = google_ai::stream_generate_content(
280 http_client.as_ref(),
281 &api_url,
282 &api_key,
283 request,
284 );
285 request.await.context("failed to stream completion")
286 }
287 .boxed()
288 }
289}
290
291impl LanguageModel for GoogleLanguageModel {
292 fn id(&self) -> LanguageModelId {
293 self.id.clone()
294 }
295
296 fn name(&self) -> LanguageModelName {
297 LanguageModelName::from(self.model.display_name().to_string())
298 }
299
300 fn provider_id(&self) -> LanguageModelProviderId {
301 PROVIDER_ID
302 }
303
304 fn provider_name(&self) -> LanguageModelProviderName {
305 PROVIDER_NAME
306 }
307
308 fn supports_tools(&self) -> bool {
309 self.model.supports_tools()
310 }
311
312 fn supports_images(&self) -> bool {
313 self.model.supports_images()
314 }
315
316 fn supports_tool_choice(&self, choice: LanguageModelToolChoice) -> bool {
317 match choice {
318 LanguageModelToolChoice::Auto
319 | LanguageModelToolChoice::Any
320 | LanguageModelToolChoice::None => true,
321 }
322 }
323
324 fn tool_input_format(&self) -> LanguageModelToolSchemaFormat {
325 LanguageModelToolSchemaFormat::JsonSchemaSubset
326 }
327
328 fn telemetry_id(&self) -> String {
329 format!("google/{}", self.model.request_id())
330 }
331
332 fn max_token_count(&self) -> u64 {
333 self.model.max_token_count()
334 }
335
336 fn max_output_tokens(&self) -> Option<u64> {
337 self.model.max_output_tokens()
338 }
339
340 fn count_tokens(
341 &self,
342 request: LanguageModelRequest,
343 cx: &App,
344 ) -> BoxFuture<'static, Result<u64>> {
345 let model_id = self.model.request_id().to_string();
346 let request = into_google(request, model_id, self.model.mode());
347 let http_client = self.http_client.clone();
348 let api_url = GoogleLanguageModelProvider::api_url(cx);
349 let api_key = self.state.read(cx).api_key_state.key(&api_url);
350
351 async move {
352 let Some(api_key) = api_key else {
353 return Err(LanguageModelCompletionError::NoApiKey {
354 provider: PROVIDER_NAME,
355 }
356 .into());
357 };
358 let response = google_ai::count_tokens(
359 http_client.as_ref(),
360 &api_url,
361 &api_key,
362 google_ai::CountTokensRequest {
363 generate_content_request: request,
364 },
365 )
366 .await?;
367 Ok(response.total_tokens)
368 }
369 .boxed()
370 }
371
372 fn stream_completion(
373 &self,
374 request: LanguageModelRequest,
375 cx: &AsyncApp,
376 ) -> BoxFuture<
377 'static,
378 Result<
379 futures::stream::BoxStream<
380 'static,
381 Result<LanguageModelCompletionEvent, LanguageModelCompletionError>,
382 >,
383 LanguageModelCompletionError,
384 >,
385 > {
386 let request = into_google(
387 request,
388 self.model.request_id().to_string(),
389 self.model.mode(),
390 );
391 let request = self.stream_completion(request, cx);
392 let future = self.request_limiter.stream(async move {
393 let response = request.await.map_err(LanguageModelCompletionError::from)?;
394 Ok(GoogleEventMapper::new().map_stream(response))
395 });
396 async move { Ok(future.await?.boxed()) }.boxed()
397 }
398}
399
400pub fn into_google(
401 mut request: LanguageModelRequest,
402 model_id: String,
403 mode: GoogleModelMode,
404) -> google_ai::GenerateContentRequest {
405 fn map_content(content: Vec<MessageContent>) -> Vec<Part> {
406 content
407 .into_iter()
408 .flat_map(|content| match content {
409 language_model::MessageContent::Text(text) => {
410 if !text.is_empty() {
411 vec![Part::TextPart(google_ai::TextPart { text })]
412 } else {
413 vec![]
414 }
415 }
416 language_model::MessageContent::Thinking {
417 text: _,
418 signature: Some(signature),
419 } => {
420 if !signature.is_empty() {
421 vec![Part::ThoughtPart(google_ai::ThoughtPart {
422 thought: true,
423 thought_signature: signature,
424 })]
425 } else {
426 vec![]
427 }
428 }
429 language_model::MessageContent::Thinking { .. } => {
430 vec![]
431 }
432 language_model::MessageContent::RedactedThinking(_) => vec![],
433 language_model::MessageContent::Image(image) => {
434 vec![Part::InlineDataPart(google_ai::InlineDataPart {
435 inline_data: google_ai::GenerativeContentBlob {
436 mime_type: "image/png".to_string(),
437 data: image.source.to_string(),
438 },
439 })]
440 }
441 language_model::MessageContent::ToolUse(tool_use) => {
442 vec![Part::FunctionCallPart(google_ai::FunctionCallPart {
443 function_call: google_ai::FunctionCall {
444 name: tool_use.name.to_string(),
445 args: tool_use.input,
446 },
447 })]
448 }
449 language_model::MessageContent::ToolResult(tool_result) => {
450 match tool_result.content {
451 language_model::LanguageModelToolResultContent::Text(text) => {
452 vec![Part::FunctionResponsePart(
453 google_ai::FunctionResponsePart {
454 function_response: google_ai::FunctionResponse {
455 name: tool_result.tool_name.to_string(),
456 // The API expects a valid JSON object
457 response: serde_json::json!({
458 "output": text
459 }),
460 },
461 },
462 )]
463 }
464 language_model::LanguageModelToolResultContent::Image(image) => {
465 vec![
466 Part::FunctionResponsePart(google_ai::FunctionResponsePart {
467 function_response: google_ai::FunctionResponse {
468 name: tool_result.tool_name.to_string(),
469 // The API expects a valid JSON object
470 response: serde_json::json!({
471 "output": "Tool responded with an image"
472 }),
473 },
474 }),
475 Part::InlineDataPart(google_ai::InlineDataPart {
476 inline_data: google_ai::GenerativeContentBlob {
477 mime_type: "image/png".to_string(),
478 data: image.source.to_string(),
479 },
480 }),
481 ]
482 }
483 }
484 }
485 })
486 .collect()
487 }
488
489 let system_instructions = if request
490 .messages
491 .first()
492 .is_some_and(|msg| matches!(msg.role, Role::System))
493 {
494 let message = request.messages.remove(0);
495 Some(SystemInstruction {
496 parts: map_content(message.content),
497 })
498 } else {
499 None
500 };
501
502 google_ai::GenerateContentRequest {
503 model: google_ai::ModelName { model_id },
504 system_instruction: system_instructions,
505 contents: request
506 .messages
507 .into_iter()
508 .filter_map(|message| {
509 let parts = map_content(message.content);
510 if parts.is_empty() {
511 None
512 } else {
513 Some(google_ai::Content {
514 parts,
515 role: match message.role {
516 Role::User => google_ai::Role::User,
517 Role::Assistant => google_ai::Role::Model,
518 Role::System => google_ai::Role::User, // Google AI doesn't have a system role
519 },
520 })
521 }
522 })
523 .collect(),
524 generation_config: Some(google_ai::GenerationConfig {
525 candidate_count: Some(1),
526 stop_sequences: Some(request.stop),
527 max_output_tokens: None,
528 temperature: request.temperature.map(|t| t as f64).or(Some(1.0)),
529 thinking_config: match (request.thinking_allowed, mode) {
530 (true, GoogleModelMode::Thinking { budget_tokens }) => {
531 budget_tokens.map(|thinking_budget| ThinkingConfig { thinking_budget })
532 }
533 _ => None,
534 },
535 top_p: None,
536 top_k: None,
537 }),
538 safety_settings: None,
539 tools: (!request.tools.is_empty()).then(|| {
540 vec![google_ai::Tool {
541 function_declarations: request
542 .tools
543 .into_iter()
544 .map(|tool| FunctionDeclaration {
545 name: tool.name,
546 description: tool.description,
547 parameters: tool.input_schema,
548 })
549 .collect(),
550 }]
551 }),
552 tool_config: request.tool_choice.map(|choice| google_ai::ToolConfig {
553 function_calling_config: google_ai::FunctionCallingConfig {
554 mode: match choice {
555 LanguageModelToolChoice::Auto => google_ai::FunctionCallingMode::Auto,
556 LanguageModelToolChoice::Any => google_ai::FunctionCallingMode::Any,
557 LanguageModelToolChoice::None => google_ai::FunctionCallingMode::None,
558 },
559 allowed_function_names: None,
560 },
561 }),
562 }
563}
564
565pub struct GoogleEventMapper {
566 usage: UsageMetadata,
567 stop_reason: StopReason,
568}
569
570impl GoogleEventMapper {
571 pub fn new() -> Self {
572 Self {
573 usage: UsageMetadata::default(),
574 stop_reason: StopReason::EndTurn,
575 }
576 }
577
578 pub fn map_stream(
579 mut self,
580 events: Pin<Box<dyn Send + Stream<Item = Result<GenerateContentResponse>>>>,
581 ) -> impl Stream<Item = Result<LanguageModelCompletionEvent, LanguageModelCompletionError>>
582 {
583 events
584 .map(Some)
585 .chain(futures::stream::once(async { None }))
586 .flat_map(move |event| {
587 futures::stream::iter(match event {
588 Some(Ok(event)) => self.map_event(event),
589 Some(Err(error)) => {
590 vec![Err(LanguageModelCompletionError::from(error))]
591 }
592 None => vec![Ok(LanguageModelCompletionEvent::Stop(self.stop_reason))],
593 })
594 })
595 }
596
597 pub fn map_event(
598 &mut self,
599 event: GenerateContentResponse,
600 ) -> Vec<Result<LanguageModelCompletionEvent, LanguageModelCompletionError>> {
601 static TOOL_CALL_COUNTER: AtomicU64 = AtomicU64::new(0);
602
603 let mut events: Vec<_> = Vec::new();
604 let mut wants_to_use_tool = false;
605 if let Some(usage_metadata) = event.usage_metadata {
606 update_usage(&mut self.usage, &usage_metadata);
607 events.push(Ok(LanguageModelCompletionEvent::UsageUpdate(
608 convert_usage(&self.usage),
609 )))
610 }
611
612 if let Some(prompt_feedback) = event.prompt_feedback
613 && let Some(block_reason) = prompt_feedback.block_reason.as_deref()
614 {
615 self.stop_reason = match block_reason {
616 "SAFETY" | "OTHER" | "BLOCKLIST" | "PROHIBITED_CONTENT" | "IMAGE_SAFETY" => {
617 StopReason::Refusal
618 }
619 _ => {
620 log::error!("Unexpected Google block_reason: {block_reason}");
621 StopReason::Refusal
622 }
623 };
624 events.push(Ok(LanguageModelCompletionEvent::Stop(self.stop_reason)));
625
626 return events;
627 }
628
629 if let Some(candidates) = event.candidates {
630 for candidate in candidates {
631 if let Some(finish_reason) = candidate.finish_reason.as_deref() {
632 self.stop_reason = match finish_reason {
633 "STOP" => StopReason::EndTurn,
634 "MAX_TOKENS" => StopReason::MaxTokens,
635 _ => {
636 log::error!("Unexpected google finish_reason: {finish_reason}");
637 StopReason::EndTurn
638 }
639 };
640 }
641 candidate
642 .content
643 .parts
644 .into_iter()
645 .for_each(|part| match part {
646 Part::TextPart(text_part) => {
647 events.push(Ok(LanguageModelCompletionEvent::Text(text_part.text)))
648 }
649 Part::InlineDataPart(_) => {}
650 Part::FunctionCallPart(function_call_part) => {
651 wants_to_use_tool = true;
652 let name: Arc<str> = function_call_part.function_call.name.into();
653 let next_tool_id =
654 TOOL_CALL_COUNTER.fetch_add(1, atomic::Ordering::SeqCst);
655 let id: LanguageModelToolUseId =
656 format!("{}-{}", name, next_tool_id).into();
657
658 events.push(Ok(LanguageModelCompletionEvent::ToolUse(
659 LanguageModelToolUse {
660 id,
661 name,
662 is_input_complete: true,
663 raw_input: function_call_part.function_call.args.to_string(),
664 input: function_call_part.function_call.args,
665 },
666 )));
667 }
668 Part::FunctionResponsePart(_) => {}
669 Part::ThoughtPart(part) => {
670 events.push(Ok(LanguageModelCompletionEvent::Thinking {
671 text: "(Encrypted thought)".to_string(), // TODO: Can we populate this from thought summaries?
672 signature: Some(part.thought_signature),
673 }));
674 }
675 });
676 }
677 }
678
679 // Even when Gemini wants to use a Tool, the API
680 // responds with `finish_reason: STOP`
681 if wants_to_use_tool {
682 self.stop_reason = StopReason::ToolUse;
683 events.push(Ok(LanguageModelCompletionEvent::Stop(StopReason::ToolUse)));
684 }
685 events
686 }
687}
688
689pub fn count_google_tokens(
690 request: LanguageModelRequest,
691 cx: &App,
692) -> BoxFuture<'static, Result<u64>> {
693 // We couldn't use the GoogleLanguageModelProvider to count tokens because the github copilot doesn't have the access to google_ai directly.
694 // So we have to use tokenizer from tiktoken_rs to count tokens.
695 cx.background_spawn(async move {
696 let messages = request
697 .messages
698 .into_iter()
699 .map(|message| tiktoken_rs::ChatCompletionRequestMessage {
700 role: match message.role {
701 Role::User => "user".into(),
702 Role::Assistant => "assistant".into(),
703 Role::System => "system".into(),
704 },
705 content: Some(message.string_contents()),
706 name: None,
707 function_call: None,
708 })
709 .collect::<Vec<_>>();
710
711 // Tiktoken doesn't yet support these models, so we manually use the
712 // same tokenizer as GPT-4.
713 tiktoken_rs::num_tokens_from_messages("gpt-4", &messages).map(|tokens| tokens as u64)
714 })
715 .boxed()
716}
717
718const fn update_usage(usage: &mut UsageMetadata, new: &UsageMetadata) {
719 if let Some(prompt_token_count) = new.prompt_token_count {
720 usage.prompt_token_count = Some(prompt_token_count);
721 }
722 if let Some(cached_content_token_count) = new.cached_content_token_count {
723 usage.cached_content_token_count = Some(cached_content_token_count);
724 }
725 if let Some(candidates_token_count) = new.candidates_token_count {
726 usage.candidates_token_count = Some(candidates_token_count);
727 }
728 if let Some(tool_use_prompt_token_count) = new.tool_use_prompt_token_count {
729 usage.tool_use_prompt_token_count = Some(tool_use_prompt_token_count);
730 }
731 if let Some(thoughts_token_count) = new.thoughts_token_count {
732 usage.thoughts_token_count = Some(thoughts_token_count);
733 }
734 if let Some(total_token_count) = new.total_token_count {
735 usage.total_token_count = Some(total_token_count);
736 }
737}
738
739fn convert_usage(usage: &UsageMetadata) -> language_model::TokenUsage {
740 let prompt_tokens = usage.prompt_token_count.unwrap_or(0);
741 let cached_tokens = usage.cached_content_token_count.unwrap_or(0);
742 let input_tokens = prompt_tokens - cached_tokens;
743 let output_tokens = usage.candidates_token_count.unwrap_or(0);
744
745 language_model::TokenUsage {
746 input_tokens,
747 output_tokens,
748 cache_read_input_tokens: cached_tokens,
749 cache_creation_input_tokens: 0,
750 }
751}
752
753struct ConfigurationView {
754 api_key_editor: Entity<SingleLineInput>,
755 state: Entity<State>,
756 target_agent: language_model::ConfigurationViewTargetAgent,
757 load_credentials_task: Option<Task<()>>,
758}
759
760impl ConfigurationView {
761 fn new(
762 state: Entity<State>,
763 target_agent: language_model::ConfigurationViewTargetAgent,
764 window: &mut Window,
765 cx: &mut Context<Self>,
766 ) -> Self {
767 cx.observe(&state, |_, _, cx| {
768 cx.notify();
769 })
770 .detach();
771
772 let load_credentials_task = Some(cx.spawn_in(window, {
773 let state = state.clone();
774 async move |this, cx| {
775 if let Some(task) = state
776 .update(cx, |state, cx| state.authenticate(cx))
777 .log_err()
778 {
779 // We don't log an error, because "not signed in" is also an error.
780 let _ = task.await;
781 }
782 this.update(cx, |this, cx| {
783 this.load_credentials_task = None;
784 cx.notify();
785 })
786 .log_err();
787 }
788 }));
789
790 Self {
791 api_key_editor: cx.new(|cx| SingleLineInput::new(window, cx, "AIzaSy...")),
792 target_agent,
793 state,
794 load_credentials_task,
795 }
796 }
797
798 fn save_api_key(&mut self, _: &menu::Confirm, window: &mut Window, cx: &mut Context<Self>) {
799 let api_key = self.api_key_editor.read(cx).text(cx).trim().to_string();
800 if api_key.is_empty() {
801 return;
802 }
803
804 // url changes can cause the editor to be displayed again
805 self.api_key_editor
806 .update(cx, |editor, cx| editor.set_text("", window, cx));
807
808 let state = self.state.clone();
809 cx.spawn_in(window, async move |_, cx| {
810 state
811 .update(cx, |state, cx| state.set_api_key(Some(api_key), cx))?
812 .await
813 })
814 .detach_and_log_err(cx);
815 }
816
817 fn reset_api_key(&mut self, window: &mut Window, cx: &mut Context<Self>) {
818 self.api_key_editor
819 .update(cx, |editor, cx| editor.set_text("", window, cx));
820
821 let state = self.state.clone();
822 cx.spawn_in(window, async move |_, cx| {
823 state
824 .update(cx, |state, cx| state.set_api_key(None, cx))?
825 .await
826 })
827 .detach_and_log_err(cx);
828 }
829
830 fn should_render_editor(&self, cx: &mut Context<Self>) -> bool {
831 !self.state.read(cx).is_authenticated()
832 }
833}
834
835impl Render for ConfigurationView {
836 fn render(&mut self, _: &mut Window, cx: &mut Context<Self>) -> impl IntoElement {
837 let env_var_set = self.state.read(cx).api_key_state.is_from_env_var();
838
839 if self.load_credentials_task.is_some() {
840 div().child(Label::new("Loading credentials...")).into_any()
841 } else if self.should_render_editor(cx) {
842 v_flex()
843 .size_full()
844 .on_action(cx.listener(Self::save_api_key))
845 .child(Label::new(format!("To use {}, you need to add an API key. Follow these steps:", match &self.target_agent {
846 ConfigurationViewTargetAgent::ZedAgent => "Zed's agent with Google AI".into(),
847 ConfigurationViewTargetAgent::Other(agent) => agent.clone(),
848 })))
849 .child(
850 List::new()
851 .child(InstructionListItem::new(
852 "Create one by visiting",
853 Some("Google AI's console"),
854 Some("https://aistudio.google.com/app/apikey"),
855 ))
856 .child(InstructionListItem::text_only(
857 "Paste your API key below and hit enter to start using the assistant",
858 )),
859 )
860 .child(self.api_key_editor.clone())
861 .child(
862 Label::new(
863 format!("You can also assign the {GEMINI_API_KEY_VAR_NAME} environment variable and restart Zed."),
864 )
865 .size(LabelSize::Small).color(Color::Muted),
866 )
867 .into_any()
868 } else {
869 h_flex()
870 .mt_1()
871 .p_1()
872 .justify_between()
873 .rounded_md()
874 .border_1()
875 .border_color(cx.theme().colors().border)
876 .bg(cx.theme().colors().background)
877 .child(
878 h_flex()
879 .gap_1()
880 .child(Icon::new(IconName::Check).color(Color::Success))
881 .child(Label::new(if env_var_set {
882 format!("API key set in {} environment variable", API_KEY_ENV_VAR.name)
883 } else {
884 let api_url = GoogleLanguageModelProvider::api_url(cx);
885 if api_url == google_ai::API_URL {
886 "API key configured".to_string()
887 } else {
888 format!("API key configured for {}", truncate_and_trailoff(&api_url, 32))
889 }
890 })),
891 )
892 .child(
893 Button::new("reset-key", "Reset Key")
894 .label_size(LabelSize::Small)
895 .icon(Some(IconName::Trash))
896 .icon_size(IconSize::Small)
897 .icon_position(IconPosition::Start)
898 .disabled(env_var_set)
899 .when(env_var_set, |this| {
900 this.tooltip(Tooltip::text(format!("To reset your API key, make sure {GEMINI_API_KEY_VAR_NAME} and {GOOGLE_AI_API_KEY_VAR_NAME} environment variables are unset.")))
901 })
902 .on_click(cx.listener(|this, _, window, cx| this.reset_api_key(window, cx))),
903 )
904 .into_any()
905 }
906 }
907}