Revert "Add context size picker for Claude Agent models" (#320614)

Revert "Add context size picker for Claude Agent models (#320166)"

This reverts commit 99d6bf7676.

extensions/copilot/src/extension/chatSessions/claude/common/claudeSessionStateService.ts

@@ -24,7 +24,6 @@ export interface SessionState {
 	folderInfo: ClaudeFolderInfo | undefined;
 	usageHandler: UsageHandler | undefined;
 	reasoningEffort: EffortLevel | undefined;
-	contextSize: number | undefined;
 	traceContext: TraceContext | undefined;
 	turnId: string | undefined;
 }
@@ -107,18 +106,6 @@ export interface IClaudeSessionStateService {
 	 */
 	setReasoningEffortForSession(sessionId: string, effort: EffortLevel | undefined): void;
 
-	/**
-	 * Gets the context size for a session (user's per-request selection from the model picker).
-	 * When set, the proxy reports this value as the endpoint's modelMaxPromptTokens so internal
-	 * accounting reflects the chosen tier.
-	 */
-	getContextSizeForSession(sessionId: string): number | undefined;
-
-	/**
-	 * Sets the context size for a session.
-	 */
-	setContextSizeForSession(sessionId: string, contextSize: number | undefined): void;
-
 	/**
 	 * Gets the OTel trace context for a session (used to parent chat spans to invoke_agent).
 	 */

extensions/copilot/src/extension/chatSessions/claude/node/claudeCodeModels.ts

@@ -8,7 +8,7 @@ import type * as vscode from 'vscode';
 import { IEndpointProvider } from '../../../../platform/endpoint/common/endpointProvider';
 import { ILogService } from '../../../../platform/log/common/logService';
 import { IChatEndpoint } from '../../../../platform/networking/common/networking';
-import { formatPricingLabel, formatTokenCount, getModelCapabilitiesDescription } from '../../../conversation/common/languageModelAccess';
+import { formatPricingLabel, getModelCapabilitiesDescription } from '../../../conversation/common/languageModelAccess';
 import { createServiceIdentifier } from '../../../../util/common/services';
 import { Emitter } from '../../../../util/vs/base/common/event';
 import { Disposable } from '../../../../util/vs/base/common/lifecycle';
@@ -17,7 +17,6 @@ import { tryParseClaudeModelId } from './claudeModelId';
 import type { EffortLevel } from '@anthropic-ai/claude-agent-sdk';
 
 export const CLAUDE_REASONING_EFFORT_PROPERTY = 'reasoningEffort';
-export const CLAUDE_CONTEXT_SIZE_PROPERTY = 'contextSize';
 
 export interface IClaudeCodeModels {
 	readonly _serviceBrand: undefined;
@@ -215,57 +214,33 @@ export function pickReasoningEffort(endpoint: IChatEndpoint | undefined, request
 }
 
 function buildConfigurationSchema(endpoint: IChatEndpoint): vscode.LanguageModelConfigurationSchema | undefined {
-	const properties: Record<string, NonNullable<vscode.LanguageModelConfigurationSchema['properties']>[string]> = {};
-
-	// Thinking effort
 	const effortLevels = endpoint.supportsReasoningEffort?.filter(
 		(level): level is typeof SUPPORTED_EFFORT_LEVELS[number] =>
 			(SUPPORTED_EFFORT_LEVELS as readonly string[]).includes(level)
 	);
-	if (effortLevels && effortLevels.length > 0) {
-		const defaultEffort = effortLevels.includes('high') ? 'high' : undefined;
-		properties[CLAUDE_REASONING_EFFORT_PROPERTY] = {
-			type: 'string',
-			title: l10n.t('Thinking Effort'),
-			enum: effortLevels,
-			enumItemLabels: effortLevels.map(level => level.charAt(0).toUpperCase() + level.slice(1)),
-			enumDescriptions: effortLevels.map(level => {
-				switch (level) {
-					case 'low': return l10n.t('Faster responses with less reasoning');
-					case 'medium': return l10n.t('Balanced reasoning and speed');
-					case 'high': return l10n.t('Greater reasoning depth but slower');
-				}
-			}),
-			default: defaultEffort,
-			group: 'navigation',
-		};
+	if (!effortLevels) {
+		return;
 	}
 
-	// Context size — only when CAPI provides a default context max, indicating
-	// a meaningful distinction between default and long context tiers.
-	const pricing = endpoint.tokenPricing;
-	const defaultContextMax = pricing?.default.contextMax;
-	const fullMax = endpoint.modelMaxPromptTokens;
-	if (defaultContextMax && defaultContextMax < fullMax) {
-		const hasLongContextSurcharge = !!pricing?.longContext;
-		properties[CLAUDE_CONTEXT_SIZE_PROPERTY] = {
-			type: 'number',
-			title: l10n.t('Context Size'),
-			enum: [defaultContextMax, fullMax],
-			enumItemLabels: [formatTokenCount(defaultContextMax), formatTokenCount(fullMax)],
-			enumDescriptions: [
-				l10n.t('Default'),
-				hasLongContextSurcharge
-					? l10n.t('Longer sessions')
-					: l10n.t('Longer sessions without compaction'),
-			],
-			default: defaultContextMax,
-			group: 'tokens',
-		};
-	}
+	const defaultEffort = effortLevels.includes('high') ? 'high' : undefined;
 
-	if (Object.keys(properties).length === 0) {
-		return undefined;
-	}
-	return { properties };
+	return {
+		properties: {
+			[CLAUDE_REASONING_EFFORT_PROPERTY]: {
+				type: 'string',
+				title: l10n.t('Thinking Effort'),
+				enum: effortLevels,
+				enumItemLabels: effortLevels.map(level => level.charAt(0).toUpperCase() + level.slice(1)),
+				enumDescriptions: effortLevels.map(level => {
+					switch (level) {
+						case 'low': return l10n.t('Faster responses with less reasoning');
+						case 'medium': return l10n.t('Balanced reasoning and speed');
+						case 'high': return l10n.t('Greater reasoning depth but slower');
+					}
+				}),
+				default: defaultEffort,
+				group: 'navigation',
+			}
+		}
+	};
 }

extensions/copilot/src/extension/chatSessions/claude/node/claudeLanguageModelServer.ts

@@ -191,17 +191,6 @@ export class ClaudeLanguageModelServer extends Disposable {
 				tokenSource.cancel();
 			});
 
-			// If the user picked a context tier in the model picker, surface that
-			// value directly as the prompt token budget. CAPI's `tokenPricing.default.contextMax`
-			// is documented as a prompt-token limit (matching `endpoint.modelMaxPromptTokens`),
-			// and CAPI bills the tier based on actual prompt size, so no other signaling is
-			// required. Clamp to the default budget so the override never lowers it.
-			const sessionContextSize = sessionId ? this.sessionStateService.getContextSizeForSession(sessionId) : undefined;
-			const defaultPromptBudget = DEFAULT_MAX_TOKENS - DEFAULT_MAX_OUTPUT_TOKENS;
-			const modelMaxPromptTokens = sessionContextSize !== undefined
-				? Math.max(defaultPromptBudget, sessionContextSize)
-				: defaultPromptBudget;
-
 			const endpointRequestBody = requestBody as IEndpointBody;
 			const streamingEndpoint = this.instantiationService.createInstance(
 				ClaudeStreamingPassThroughEndpoint,
@@ -211,7 +200,7 @@ export class ClaudeLanguageModelServer extends Disposable {
 				headers,
 				'vscode_claude_code',
 				{
-					modelMaxPromptTokens,
+					modelMaxPromptTokens: DEFAULT_MAX_TOKENS - DEFAULT_MAX_OUTPUT_TOKENS,
 					maxOutputTokens: DEFAULT_MAX_OUTPUT_TOKENS
 				},
 				sessionId

extensions/copilot/src/extension/chatSessions/claude/node/claudeSessionStateService.ts

@@ -47,7 +47,6 @@ export class ClaudeSessionStateService extends Disposable implements IClaudeSess
 			folderInfo: existing?.folderInfo,
 			usageHandler: existing?.usageHandler,
 			reasoningEffort: existing?.reasoningEffort,
-			contextSize: existing?.contextSize,
 			traceContext: existing?.traceContext,
 			turnId: existing?.turnId,
 		});
@@ -70,7 +69,6 @@ export class ClaudeSessionStateService extends Disposable implements IClaudeSess
 			folderInfo: existing?.folderInfo,
 			usageHandler: existing?.usageHandler,
 			reasoningEffort: existing?.reasoningEffort,
-			contextSize: existing?.contextSize,
 			traceContext: existing?.traceContext,
 			turnId: existing?.turnId,
 		});
@@ -90,7 +88,6 @@ export class ClaudeSessionStateService extends Disposable implements IClaudeSess
 			folderInfo: existing?.folderInfo,
 			usageHandler: existing?.usageHandler,
 			reasoningEffort: existing?.reasoningEffort,
-			contextSize: existing?.contextSize,
 			traceContext: existing?.traceContext,
 			turnId: existing?.turnId,
 		});
@@ -112,7 +109,6 @@ export class ClaudeSessionStateService extends Disposable implements IClaudeSess
 			folderInfo,
 			usageHandler: existing?.usageHandler,
 			reasoningEffort: existing?.reasoningEffort,
-			contextSize: existing?.contextSize,
 			traceContext: existing?.traceContext,
 			turnId: existing?.turnId,
 		});
@@ -132,7 +128,6 @@ export class ClaudeSessionStateService extends Disposable implements IClaudeSess
 			folderInfo: existing?.folderInfo,
 			usageHandler: handler,
 			reasoningEffort: existing?.reasoningEffort,
-			contextSize: existing?.contextSize,
 			traceContext: existing?.traceContext,
 			turnId: existing?.turnId,
 		});
@@ -154,29 +149,6 @@ export class ClaudeSessionStateService extends Disposable implements IClaudeSess
 			folderInfo: existing?.folderInfo,
 			usageHandler: existing?.usageHandler,
 			reasoningEffort: effort,
-			contextSize: existing?.contextSize,
-			traceContext: existing?.traceContext,
-			turnId: existing?.turnId,
-		});
-	}
-
-	getContextSizeForSession(sessionId: string): number | undefined {
-		return this._sessionState.get(sessionId)?.contextSize;
-	}
-
-	setContextSizeForSession(sessionId: string, contextSize: number | undefined): void {
-		const existing = this._sessionState.get(sessionId);
-		if (existing?.contextSize === contextSize) {
-			return;
-		}
-		this._sessionState.set(sessionId, {
-			modelId: existing?.modelId,
-			permissionMode: existing?.permissionMode ?? 'acceptEdits',
-			capturingToken: existing?.capturingToken,
-			folderInfo: existing?.folderInfo,
-			usageHandler: existing?.usageHandler,
-			reasoningEffort: existing?.reasoningEffort,
-			contextSize,
 			traceContext: existing?.traceContext,
 			turnId: existing?.turnId,
 		});
@@ -195,7 +167,6 @@ export class ClaudeSessionStateService extends Disposable implements IClaudeSess
 			folderInfo: existing?.folderInfo,
 			usageHandler: existing?.usageHandler,
 			reasoningEffort: existing?.reasoningEffort,
-			contextSize: existing?.contextSize,
 			traceContext,
 			turnId: existing?.turnId,
 		});
@@ -214,7 +185,6 @@ export class ClaudeSessionStateService extends Disposable implements IClaudeSess
 			folderInfo: existing?.folderInfo,
 			usageHandler: existing?.usageHandler,
 			reasoningEffort: existing?.reasoningEffort,
-			contextSize: existing?.contextSize,
 			traceContext: existing?.traceContext,
 			turnId,
 		});

extensions/copilot/src/extension/chatSessions/claude/node/test/claudeCodeModels.spec.ts

@@ -6,7 +6,7 @@
 import { afterEach, describe, expect, it } from 'vitest';
 import type * as vscode from 'vscode';
 import { IEndpointProvider } from '../../../../../platform/endpoint/common/endpointProvider';
-import { IChatEndpoint, IChatEndpointTokenPricing } from '../../../../../platform/networking/common/networking';
+import { IChatEndpoint } from '../../../../../platform/networking/common/networking';
 import { Emitter } from '../../../../../util/vs/base/common/event';
 import { DisposableStore } from '../../../../../util/vs/base/common/lifecycle';
 import { IInstantiationService } from '../../../../../util/vs/platform/instantiation/common/instantiation';
@@ -26,8 +26,6 @@ function createMockEndpoint(overrides: {
 	apiType?: string;
 	modelProvider?: string;
 	supportsReasoningEffort?: string[];
-	tokenPricing?: IChatEndpointTokenPricing;
-	modelMaxPromptTokens?: number;
 }): IChatEndpoint {
 	const isAnthropic = overrides.modelProvider === undefined || overrides.modelProvider === 'Anthropic';
 	return {
@@ -49,8 +47,7 @@ function createMockEndpoint(overrides: {
 		isFallback: false,
 		policy: 'enabled',
 		urlOrRequestMetadata: 'mock://endpoint',
-		modelMaxPromptTokens: overrides.modelMaxPromptTokens ?? 128000,
-		tokenPricing: overrides.tokenPricing,
+		modelMaxPromptTokens: 128000,
 		tokenizer: 'cl100k_base',
 		acquireTokenizer: () => ({ encode: () => [], free: () => { } }) as any,
 		processResponseFromChatEndpoint: () => Promise.resolve({} as any),
@@ -328,110 +325,6 @@ describe('ClaudeCodeModels', () => {
 			expect(schema.properties?.['reasoningEffort'].enum).toEqual(['high']);
 			expect(schema.properties!['reasoningEffort'].default).toBe('high');
 		});
-
-		it('includes contextSize when endpoint pricing exposes a default context max below modelMaxPromptTokens', async () => {
-			const { service } = createServiceWithRefreshableEndpoints([
-				createMockEndpoint({
-					model: 'claude-sonnet-4-model',
-					name: 'Claude Sonnet 4',
-					family: 'claude-sonnet-4',
-					modelMaxPromptTokens: 1_000_000,
-					tokenPricing: {
-						default: { inputPrice: 3, outputPrice: 15, cacheReadTokenPrice: 0.3, contextMax: 200_000 },
-						longContext: { inputPrice: 6, outputPrice: 22.5, cacheReadTokenPrice: 0.6 },
-					},
-				}),
-			]);
-			const { lm, getCapturedProvider } = createMockLm();
-
-			const info = await getProviderInfo(service, lm, getCapturedProvider);
-			const schema = info[0].configurationSchema!;
-			expect(schema.properties?.['contextSize']).toEqual({
-				type: 'number',
-				title: 'Context Size',
-				enum: [200_000, 1_000_000],
-				enumItemLabels: ['200K', '1M'],
-				enumDescriptions: ['Default', 'Longer sessions'],
-				default: 200_000,
-				group: 'tokens',
-			});
-		});
-
-		it('omits contextSize when pricing has no default context max', async () => {
-			const { service } = createServiceWithRefreshableEndpoints([
-				createMockEndpoint({
-					model: 'claude-sonnet-4-model',
-					name: 'Claude Sonnet 4',
-					family: 'claude-sonnet-4',
-					modelMaxPromptTokens: 1_000_000,
-					tokenPricing: {
-						default: { inputPrice: 3, outputPrice: 15, cacheReadTokenPrice: 0.3 },
-					},
-				}),
-			]);
-			const { lm, getCapturedProvider } = createMockLm();
-
-			const info = await getProviderInfo(service, lm, getCapturedProvider);
-			expect(info[0].configurationSchema).toBeUndefined();
-		});
-
-		it('omits contextSize when default context max equals modelMaxPromptTokens', async () => {
-			const { service } = createServiceWithRefreshableEndpoints([
-				createMockEndpoint({
-					model: 'claude-sonnet-4-model',
-					name: 'Claude Sonnet 4',
-					family: 'claude-sonnet-4',
-					modelMaxPromptTokens: 200_000,
-					tokenPricing: {
-						default: { inputPrice: 3, outputPrice: 15, cacheReadTokenPrice: 0.3, contextMax: 200_000 },
-					},
-				}),
-			]);
-			const { lm, getCapturedProvider } = createMockLm();
-
-			const info = await getProviderInfo(service, lm, getCapturedProvider);
-			expect(info[0].configurationSchema).toBeUndefined();
-		});
-
-		it('describes the long-context option without surcharge when there is no long-context pricing tier', async () => {
-			const { service } = createServiceWithRefreshableEndpoints([
-				createMockEndpoint({
-					model: 'claude-sonnet-4-model',
-					name: 'Claude Sonnet 4',
-					family: 'claude-sonnet-4',
-					modelMaxPromptTokens: 1_000_000,
-					tokenPricing: {
-						default: { inputPrice: 3, outputPrice: 15, cacheReadTokenPrice: 0.3, contextMax: 200_000 },
-					},
-				}),
-			]);
-			const { lm, getCapturedProvider } = createMockLm();
-
-			const info = await getProviderInfo(service, lm, getCapturedProvider);
-			const schema = info[0].configurationSchema!;
-			expect(schema.properties?.['contextSize'].enumDescriptions).toEqual(['Default', 'Longer sessions without compaction']);
-		});
-
-		it('includes both reasoningEffort and contextSize when supported', async () => {
-			const { service } = createServiceWithRefreshableEndpoints([
-				createMockEndpoint({
-					model: 'claude-sonnet-4-model',
-					name: 'Claude Sonnet 4',
-					family: 'claude-sonnet-4',
-					supportsReasoningEffort: ['low', 'medium', 'high'],
-					modelMaxPromptTokens: 1_000_000,
-					tokenPricing: {
-						default: { inputPrice: 3, outputPrice: 15, cacheReadTokenPrice: 0.3, contextMax: 200_000 },
-						longContext: { inputPrice: 6, outputPrice: 22.5, cacheReadTokenPrice: 0.6 },
-					},
-				}),
-			]);
-			const { lm, getCapturedProvider } = createMockLm();
-
-			const info = await getProviderInfo(service, lm, getCapturedProvider);
-			const schema = info[0].configurationSchema!;
-			expect(Object.keys(schema.properties ?? {})).toEqual(['reasoningEffort', 'contextSize']);
-		});
 	});
 
 	describe('resolveEndpoint with ParsedClaudeModelId', () => {

extensions/copilot/src/extension/chatSessions/claude/node/test/claudeSessionStateService.spec.ts

@@ -268,57 +268,6 @@ describe('ClaudeSessionStateService', () => {
 		});
 	});
 
-	describe('getContextSizeForSession', () => {
-		it('should return undefined when no context size is set', () => {
-			assert.strictEqual(service.getContextSizeForSession('session-1'), undefined);
-		});
-
-		it('should return the set context size', () => {
-			service.setContextSizeForSession('session-1', 200_000);
-			assert.strictEqual(service.getContextSizeForSession('session-1'), 200_000);
-		});
-
-		it('should return different sizes for different sessions', () => {
-			service.setContextSizeForSession('session-1', 200_000);
-			service.setContextSizeForSession('session-2', 1_000_000);
-
-			assert.strictEqual(service.getContextSizeForSession('session-1'), 200_000);
-			assert.strictEqual(service.getContextSizeForSession('session-2'), 1_000_000);
-		});
-	});
-
-	describe('setContextSizeForSession', () => {
-		it('should allow setting a context size', () => {
-			service.setContextSizeForSession('session-1', 1_000_000);
-			assert.strictEqual(service.getContextSizeForSession('session-1'), 1_000_000);
-		});
-
-		it('should allow clearing a context size', () => {
-			service.setContextSizeForSession('session-1', 1_000_000);
-			service.setContextSizeForSession('session-1', undefined);
-			assert.strictEqual(service.getContextSizeForSession('session-1'), undefined);
-		});
-
-		it('should preserve other state when setting context size', () => {
-			service.setModelIdForSession('session-1', OPUS_4);
-			service.setReasoningEffortForSession('session-1', 'high');
-
-			service.setContextSizeForSession('session-1', 1_000_000);
-
-			assert.strictEqual(service.getModelIdForSession('session-1'), OPUS_4);
-			assert.strictEqual(service.getReasoningEffortForSession('session-1'), 'high');
-		});
-
-		it('should not fire onDidChangeSessionState event', () => {
-			const events: SessionStateChangeEvent[] = [];
-			service.onDidChangeSessionState(e => events.push(e));
-
-			service.setContextSizeForSession('session-1', 1_000_000);
-
-			assert.strictEqual(events.length, 0);
-		});
-	});
-
 	describe('dispose', () => {
 		it('should clear session state on dispose', () => {
 			service.setModelIdForSession('session-1', OPUS_4);

extensions/copilot/src/extension/chatSessions/vscode-node/claudeChatSessionContentProvider.ts

@@ -26,7 +26,7 @@ import { IInstantiationService } from '../../../util/vs/platform/instantiation/c
 import { ClaudeFolderInfo } from '../claude/common/claudeFolderInfo';
 import { ClaudeSessionUri } from '../claude/common/claudeSessionUri';
 import { ClaudeAgentManager } from '../claude/node/claudeCodeAgent';
-import { CLAUDE_CONTEXT_SIZE_PROPERTY, CLAUDE_REASONING_EFFORT_PROPERTY, IClaudeCodeModels, pickReasoningEffort } from '../claude/node/claudeCodeModels';
+import { CLAUDE_REASONING_EFFORT_PROPERTY, IClaudeCodeModels, pickReasoningEffort } from '../claude/node/claudeCodeModels';
 import { IClaudeCodeSdkService } from '../claude/node/claudeCodeSdkService';
 import { parseClaudeModelId } from '../claude/node/claudeModelId';
 import { IClaudeSessionStateService } from '../claude/common/claudeSessionStateService';
@@ -177,10 +177,6 @@ export class ClaudeChatSessionContentProvider extends Disposable implements vsco
 			const reasoningEffort = pickReasoningEffort(endpoint, typeof rawReasoningEffort === 'string' ? rawReasoningEffort : undefined);
 			this.sessionStateService.setReasoningEffortForSession(effectiveSessionId, reasoningEffort);
 
-			const rawContextSize = request.modelConfiguration?.[CLAUDE_CONTEXT_SIZE_PROPERTY];
-			const contextSize = typeof rawContextSize === 'number' && rawContextSize > 0 ? rawContextSize : undefined;
-			this.sessionStateService.setContextSizeForSession(effectiveSessionId, contextSize);
-
 			// Set usage handler to report token usage for context window widget
 			this.sessionStateService.setUsageHandlerForSession(effectiveSessionId, (usage) => {
 				stream.usage(usage);