feat: Enhance Claude Sonnet 4 support with 1M context window and tiered pricing (#196)

Author: magesh-presidio

.changeset/fair-buttons-chew.md

@@ -0,0 +1,5 @@
+---
+"hai-build-code-generator": patch
+---
+
+Enhanced support for Claude Sonnet 4, extending its maximum context window to 1 million tokens and enabling tiered pricing for more flexible usage models.

package-lock.json

@@ -386,7 +386,7 @@
 		"@anthropic-ai/bedrock-sdk": "^0.12.4",
 		"@anthropic-ai/sdk": "^0.37.0",
 		"@anthropic-ai/vertex-sdk": "^0.6.4",
-		"@aws-sdk/client-bedrock-runtime": "^3.758.0",
+		"@aws-sdk/client-bedrock-runtime": "^3.873.0",
 		"@bufbuild/protobuf": "^2.2.5",
 		"@google-cloud/vertexai": "^1.9.3",
 		"@google/genai": "^0.13.0",

package.json

@@ -35,6 +35,27 @@ message VsCodeLmModel {
   string id = 4;
 }
 
+// Price tier for tiered pricing models
+message PriceTier {
+  int32 token_limit = 1;  // Upper limit (inclusive) of input tokens for this price
+  double price = 2;       // Price per million tokens for this tier
+}
+
+message ThinkingConfig {
+  optional int32 max_budget = 1;                    // Max allowed thinking budget tokens
+  optional double output_price = 2;                 // Output price per million tokens when budget > 0
+  repeated PriceTier output_price_tiers = 3;        // Optional: Tiered output price when budget > 0
+}
+
+// Model tier for tiered pricing structures
+message ModelTier {
+  int32 context_window = 1;
+  optional double input_price = 2;
+  optional double output_price = 3;
+  optional double cache_writes_price = 4;
+  optional double cache_reads_price = 5;
+}
+
 // For OpenRouterCompatibleModelInfo structure in OpenRouterModels
 message OpenRouterModelInfo {
   int32 max_tokens = 1;
@@ -46,6 +67,9 @@ message OpenRouterModelInfo {
   double cache_writes_price = 7;
   double cache_reads_price = 8;
   string description = 9;
+  optional ThinkingConfig thinking_config = 10;
+  optional bool supports_global_endpoint = 11;
+  repeated ModelTier tiers = 12;
 }
 
 // Shared response message for model information

proto/models.proto

@@ -1,15 +1,23 @@
 import { Anthropic } from "@anthropic-ai/sdk"
 import { Stream as AnthropicStream } from "@anthropic-ai/sdk/streaming"
 import { withRetry } from "../retry"
-import { anthropicDefaultModelId, AnthropicModelId, anthropicModels, ApiHandlerOptions, ModelInfo } from "@shared/api"
+import { anthropicDefaultModelId, AnthropicModelId, anthropicModels, CLAUDE_SONNET_4_1M_SUFFIX, ModelInfo } from "@shared/api"
 import { ApiHandler } from "../index"
 import { ApiStream } from "../transform/stream"
 
+interface AnthropicHandlerOptions {
+	apiKey?: string
+	anthropicBaseUrl?: string
+	apiModelId?: string
+	thinkingBudgetTokens?: number
+	maxRetries?: number
+}
+
 export class AnthropicHandler implements ApiHandler {
-	private options: ApiHandlerOptions
+	private options: AnthropicHandlerOptions
 	private client: Anthropic
 
-	constructor(options: ApiHandlerOptions) {
+	constructor(options: AnthropicHandlerOptions) {
 		this.options = options
 		this.client = new Anthropic({
 			apiKey: this.options.apiKey,
@@ -22,7 +30,10 @@ export class AnthropicHandler implements ApiHandler {
 	async *createMessage(systemPrompt: string, messages: Anthropic.Messages.MessageParam[]): ApiStream {
 		const model = this.getModel()
 		let stream: AnthropicStream<Anthropic.RawMessageStreamEvent>
-		const modelId = model.id
+		const modelId = model.id.endsWith(CLAUDE_SONNET_4_1M_SUFFIX)
+			? model.id.slice(0, -CLAUDE_SONNET_4_1M_SUFFIX.length)
+			: model.id
+		const enable1mContextWindow = model.id.endsWith(CLAUDE_SONNET_4_1M_SUFFIX)
 
 		const budget_tokens = this.options.thinkingBudgetTokens || 0
 		const reasoningOn = (modelId.includes("3-7") || modelId.includes("4-")) && budget_tokens !== 0 ? true : false
@@ -95,24 +106,15 @@ export class AnthropicHandler implements ApiHandler {
 						stream: true,
 					},
 					(() => {
-						// prompt caching: https://x.com/alexalbert__/status/1823751995901272068
-						// https://github.com/anthropics/anthropic-sdk-typescript?tab=readme-ov-file#default-headers
-						// https://github.com/anthropics/anthropic-sdk-typescript/commit/c920b77fc67bd839bfeb6716ceab9d7c9bbe7393
-						switch (modelId) {
-							case "claude-sonnet-4-20250514":
-							case "claude-opus-4-20250514":
-							case "claude-3-7-sonnet-20250219":
-							case "claude-3-5-sonnet-20241022":
-							case "claude-3-5-haiku-20241022":
-							case "claude-3-opus-20240229":
-							case "claude-3-haiku-20240307":
-								return {
-									headers: {
-										"anthropic-beta": "prompt-caching-2024-07-31",
-									},
-								}
-							default:
-								return undefined
+						// 1m context window beta header
+						if (enable1mContextWindow) {
+							return {
+								headers: {
+									"anthropic-beta": "context-1m-2025-08-07",
+								},
+							}
+						} else {
+							return undefined
 						}
 					})(),
 				)