fix(ai): ensure that tool calls batched in a single turn always run in parallel

CHANGELOG.md

@@ -4,6 +4,20 @@
 
 - [Previous Changelogs](https://github.com/eclipse-theia/theia/tree/master/doc/changelogs/)
 
+## 1.76.0 - tbd
+
+- [ai-anthropic, ai-google, ai-ollama, ai-openai, ai-copilot] added rebindable `<provider>LanguageModelFactory` bindings for instantiating provider language models [#17623](https://github.com/eclipse-theia/theia/pull/17623)
+- [ai-core] executed a model turn's tool calls concurrently, including parallel agent delegations, instead of sequentially, via a new injectable `ToolCallExecutor` [#17623](https://github.com/eclipse-theia/theia/pull/17623)
+- [ai-anthropic, ai-google, ai-ollama, ai-openai, ai-copilot] the provider language model classes are now `@injectable`, transient-scoped services constructed via their `<provider>LanguageModelFactory`, so adopters can rebind them to substitute custom implementations [#17623](https://github.com/eclipse-theia/theia/pull/17623)
+- [ai-openai] added a rebindable `ChatCompletionStreamingAsyncIteratorFactory`; the chat-completion tool-call streaming iterator (used by the OpenAI and Copilot models) is now an injectable service that can be substituted [#17623](https://github.com/eclipse-theia/theia/pull/17623)
+
+<a name="breaking_changes_1.76.0">[Breaking Changes:](#breaking_changes_1.76.0)</a>
+
+- [ai-openai] `OpenAiLanguageModelsManagerImpl` no longer injects `OpenAiModelUtils` or `OpenAiResponseApiUtils` (the `openAiModelUtils` and `responseApiUtils` protected fields were removed); provider models are now constructed via the injected `OpenAiLanguageModelFactory` [#17623](https://github.com/eclipse-theia/theia/pull/17623)
+- [ai-openai, ai-copilot] `OpenAiModel.createTools()` and `CopilotLanguageModel.createTools()` now return `ChatCompletionTool[]` instead of `RunnableToolFunctionWithoutParse[]`, because the OpenAI SDK `runTools` runner is no longer used [#17623](https://github.com/eclipse-theia/theia/pull/17623)
+- [ai-copilot] removed the `protected runnerOptions` field from `CopilotLanguageModel`; its only purpose was the `maxChatCompletions` turn cap, which is gone now that the OpenAI SDK `runTools` runner is unused, so the tool loop runs until the model stops requesting tools. Subclasses that read or overrode `runnerOptions` must adapt. `OpenAiModel.runnerOptions` is retained, but its `maxChatCompletions` now bounds only the Response API path, not the Chat Completions tool loop [#17623](https://github.com/eclipse-theia/theia/pull/17623)
+- [ai-anthropic, ai-google, ai-ollama, ai-openai, ai-copilot] the provider language model classes (`AnthropicModel`, `GoogleModel`, `OllamaModel`, `OpenAiModel`, `CopilotLanguageModel`) no longer expose public constructors; they are `@injectable` and receive their configuration through an injected `<provider>ModelParams` object (a symbol) plus injected service dependencies. Instantiate them via the corresponding `<provider>LanguageModelFactory` (or the DI container) instead of `new`, and drop constructor overrides in subclasses [#17623](https://github.com/eclipse-theia/theia/pull/17623)
+
 ## 1.75.0 - 8/27/2026
 
 - [ai] added `gemini-3.7-flash` to the default models [#17947](https://github.com/eclipse-theia/theia/pull/17947) - Contributed on behalf of STMicroelectronics

packages/ai-anthropic/src/node/anthropic-backend-module.ts

@@ -14,17 +14,27 @@
 // SPDX-License-Identifier: EPL-2.0 OR GPL-2.0-only WITH Classpath-exception-2.0
 // *****************************************************************************
 
-import { ContainerModule } from '@theia/core/shared/inversify';
+import { Container, ContainerModule } from '@theia/core/shared/inversify';
 import { ANTHROPIC_LANGUAGE_MODELS_MANAGER_PATH, AnthropicLanguageModelsManager } from '../common/anthropic-language-models-manager';
 import { ConnectionHandler, PreferenceContribution, RpcConnectionHandler } from '@theia/core';
 import { AnthropicLanguageModelsManagerImpl } from './anthropic-language-models-manager-impl';
+import { AnthropicLanguageModelFactory, AnthropicModel, AnthropicModelParams } from './anthropic-language-model';
 import { ConnectionContainerModule } from '@theia/core/lib/node/messaging/connection-container-module';
 import { AnthropicPreferencesSchema } from '../common/anthropic-preferences';
 
 // We use a connection module to handle AI services separately for each frontend.
 const anthropicConnectionModule = ConnectionContainerModule.create(({ bind, bindBackendService, bindFrontendService }) => {
     bind(AnthropicLanguageModelsManagerImpl).toSelf().inSingletonScope();
     bind(AnthropicLanguageModelsManager).toService(AnthropicLanguageModelsManagerImpl);
+    bind(AnthropicModel).toSelf().inTransientScope();
+    bind(AnthropicLanguageModelFactory).toFactory<AnthropicModel, [AnthropicModelParams]>(
+        ({ container }) => params => {
+            const child = new Container();
+            child.parent = container;
+            child.bind(AnthropicModelParams).toConstantValue(params);
+            return child.get(AnthropicModel);
+        }
+    );
     bind(ConnectionHandler).toDynamicValue(ctx =>
         new RpcConnectionHandler(ANTHROPIC_LANGUAGE_MODELS_MANAGER_PATH, () => ctx.container.get(AnthropicLanguageModelsManager))
     ).inSingletonScope();