@aliou/pi-synthetic 0.13.0 → 0.13.2

package/README.md

@@ -1,3 +1,5 @@
+![banner](https://assets.aliou.me/pi-extensions/banners/pi-synthetic.png)
+
 # Pi Synthetic Extension
 
 A Pi extension that adds [Synthetic](https://synthetic.new) as a model provider, giving you access to open-source models through an OpenAI-compatible API.
@@ -162,4 +164,4 @@ This repository uses [Changesets](https://github.com/changesets/changesets) for
 - [Synthetic](https://synthetic.new)
 - [Synthetic Models](https://synthetic.new/models)
 - [Synthetic API Docs](https://dev.synthetic.new/docs/api/overview)
-- [Pi Documentation](https://buildwithpi.ai/)
+- [Pi Documentation](https://buildwithpi.ai/)

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@aliou/pi-synthetic",
-  "version": "0.13.0",
+  "version": "0.13.2",
   "license": "MIT",
   "type": "module",
   "private": false,

package/src/extensions/provider/index.ts

@@ -63,14 +63,8 @@ export function registerSyntheticProvider(pi: ExtensionAPI): void {
       Referer: "https://pi.dev",
       "X-Title": "npm:@aliou/pi-synthetic",
     },
-    models: SYNTHETIC_MODELS.map((model) => ({
-      id: model.id,
-      name: model.name,
-      reasoning: model.reasoning,
-      input: model.input,
-      cost: model.cost,
-      contextWindow: model.contextWindow,
-      maxTokens: model.maxTokens,
+    models: SYNTHETIC_MODELS.map(({ provider: _provider, ...model }) => ({
+      ...model,
       compat: {
         supportsDeveloperRole: false,
         maxTokensField: "max_tokens",

package/src/extensions/provider/models.test.ts

@@ -4,6 +4,7 @@ import { SYNTHETIC_MODELS } from "./models";
 interface ApiModel {
   id: string;
   name: string;
+  provider: string | null;
   input_modalities: string[];
   output_modalities: string[];
   context_length: number;
@@ -156,6 +157,19 @@ function compareModels(
         });
       }
     }
+
+    // Check provider
+    if (
+      apiModel.provider !== null &&
+      apiModel.provider !== hardcoded.provider
+    ) {
+      discrepancies.push({
+        model: hardcoded.id,
+        field: "provider",
+        hardcoded: hardcoded.provider,
+        api: apiModel.provider,
+      });
+    }
   }
 
   // Check for API models not in hardcoded list

package/src/extensions/provider/models.ts

@@ -2,29 +2,11 @@
 // Source: https://api.synthetic.new/openai/v1/models
 // maxTokens sourced from https://models.dev/api.json (synthetic provider)
 
-export interface SyntheticModelConfig {
-  id: string;
-  name: string;
-  reasoning: boolean;
-  input: ("text" | "image")[];
-  cost: {
-    input: number;
-    output: number;
-    cacheRead: number;
-    cacheWrite: number;
-  };
-  contextWindow: number;
-  maxTokens: number;
-  compat?: {
-    supportsDeveloperRole?: boolean;
-    supportsReasoningEffort?: boolean;
-    reasoningEffortMap?: Partial<
-      Record<"minimal" | "low" | "medium" | "high" | "xhigh", string>
-    >;
-    maxTokensField?: "max_completion_tokens" | "max_tokens";
-    requiresToolResultName?: boolean;
-    requiresMistralToolIds?: boolean;
-  };
+import type { ProviderModelConfig } from "@mariozechner/pi-coding-agent";
+
+export interface SyntheticModelConfig extends ProviderModelConfig {
+  /** Upstream backend Synthetic proxies this model through (e.g. "fireworks", "together", "synthetic"). */
+  provider: string;
 }
 
 const SYNTHETIC_REASONING_EFFORT_MAP = {
@@ -36,20 +18,21 @@ const SYNTHETIC_REASONING_EFFORT_MAP = {
 } as const;
 
 export const SYNTHETIC_MODELS: SyntheticModelConfig[] = [
-  // API: hf:zai-org/GLM-4.7 → ctx=202752 (provider: fireworks)
+  // API: hf:zai-org/GLM-4.7 → ctx=202752
   {
     id: "hf:zai-org/GLM-4.7",
     name: "zai-org/GLM-4.7",
+    provider: "synthetic",
     reasoning: true,
     compat: {
       supportsReasoningEffort: true,
       reasoningEffortMap: SYNTHETIC_REASONING_EFFORT_MAP,
     },
-    input: ["text", "image"],
+    input: ["text"],
     cost: {
-      input: 2.19,
+      input: 0.45,
       output: 2.19,
-      cacheRead: 2.19,
+      cacheRead: 0.45,
       cacheWrite: 0,
     },
     contextWindow: 202752,
@@ -59,6 +42,7 @@ export const SYNTHETIC_MODELS: SyntheticModelConfig[] = [
   {
     id: "hf:zai-org/GLM-5",
     name: "zai-org/GLM-5",
+    provider: "synthetic",
     reasoning: true,
     compat: {
       supportsReasoningEffort: true,
@@ -78,6 +62,7 @@ export const SYNTHETIC_MODELS: SyntheticModelConfig[] = [
   {
     id: "hf:zai-org/GLM-5.1",
     name: "zai-org/GLM-5.1",
+    provider: "synthetic",
     reasoning: true,
     compat: {
       supportsReasoningEffort: true,
@@ -98,6 +83,7 @@ export const SYNTHETIC_MODELS: SyntheticModelConfig[] = [
   {
     id: "hf:zai-org/GLM-4.7-Flash",
     name: "zai-org/GLM-4.7-Flash",
+    provider: "synthetic",
     reasoning: true,
     compat: {
       supportsReasoningEffort: true,
@@ -117,6 +103,7 @@ export const SYNTHETIC_MODELS: SyntheticModelConfig[] = [
   {
     id: "hf:MiniMaxAI/MiniMax-M2.1",
     name: "MiniMaxAI/MiniMax-M2.1",
+    provider: "fireworks",
     reasoning: true,
     compat: {
       supportsReasoningEffort: true,
@@ -136,6 +123,7 @@ export const SYNTHETIC_MODELS: SyntheticModelConfig[] = [
   {
     id: "hf:meta-llama/Llama-3.3-70B-Instruct",
     name: "meta-llama/Llama-3.3-70B-Instruct",
+    provider: "together",
     reasoning: false,
     input: ["text"],
     cost: {
@@ -151,6 +139,7 @@ export const SYNTHETIC_MODELS: SyntheticModelConfig[] = [
   {
     id: "hf:deepseek-ai/DeepSeek-R1-0528",
     name: "deepseek-ai/DeepSeek-R1-0528",
+    provider: "together",
     reasoning: true,
     compat: {
       supportsReasoningEffort: true,
@@ -170,6 +159,7 @@ export const SYNTHETIC_MODELS: SyntheticModelConfig[] = [
   {
     id: "hf:deepseek-ai/DeepSeek-V3.2",
     name: "deepseek-ai/DeepSeek-V3.2",
+    provider: "fireworks",
     reasoning: false,
     input: ["text"],
     cost: {
@@ -185,6 +175,7 @@ export const SYNTHETIC_MODELS: SyntheticModelConfig[] = [
   {
     id: "hf:moonshotai/Kimi-K2-Instruct-0905",
     name: "moonshotai/Kimi-K2-Instruct-0905",
+    provider: "fireworks",
     reasoning: false,
     input: ["text"],
     cost: {
@@ -200,6 +191,7 @@ export const SYNTHETIC_MODELS: SyntheticModelConfig[] = [
   {
     id: "hf:moonshotai/Kimi-K2-Thinking",
     name: "moonshotai/Kimi-K2-Thinking",
+    provider: "fireworks",
     reasoning: true,
     compat: {
       supportsReasoningEffort: true,
@@ -219,6 +211,7 @@ export const SYNTHETIC_MODELS: SyntheticModelConfig[] = [
   {
     id: "hf:openai/gpt-oss-120b",
     name: "openai/gpt-oss-120b",
+    provider: "fireworks",
     reasoning: false,
     input: ["text"],
     cost: {
@@ -234,6 +227,7 @@ export const SYNTHETIC_MODELS: SyntheticModelConfig[] = [
   {
     id: "hf:Qwen/Qwen3-Coder-480B-A35B-Instruct",
     name: "Qwen/Qwen3-Coder-480B-A35B-Instruct",
+    provider: "together",
     reasoning: true,
     compat: {
       supportsReasoningEffort: true,
@@ -253,6 +247,7 @@ export const SYNTHETIC_MODELS: SyntheticModelConfig[] = [
   {
     id: "hf:moonshotai/Kimi-K2.5",
     name: "moonshotai/Kimi-K2.5",
+    provider: "synthetic",
     reasoning: true,
     compat: {
       supportsReasoningEffort: true,
@@ -272,6 +267,7 @@ export const SYNTHETIC_MODELS: SyntheticModelConfig[] = [
   {
     id: "hf:nvidia/Kimi-K2.5-NVFP4",
     name: "nvidia/Kimi-K2.5-NVFP4",
+    provider: "synthetic",
     reasoning: true,
     compat: {
       supportsReasoningEffort: true,
@@ -291,6 +287,7 @@ export const SYNTHETIC_MODELS: SyntheticModelConfig[] = [
   {
     id: "hf:deepseek-ai/DeepSeek-V3",
     name: "deepseek-ai/DeepSeek-V3",
+    provider: "together",
     reasoning: false,
     input: ["text"],
     cost: {
@@ -306,6 +303,7 @@ export const SYNTHETIC_MODELS: SyntheticModelConfig[] = [
   {
     id: "hf:Qwen/Qwen3-235B-A22B-Thinking-2507",
     name: "Qwen/Qwen3-235B-A22B-Thinking-2507",
+    provider: "together",
     reasoning: true,
     compat: {
       supportsReasoningEffort: true,
@@ -325,6 +323,7 @@ export const SYNTHETIC_MODELS: SyntheticModelConfig[] = [
   {
     id: "hf:Qwen/Qwen3.5-397B-A17B",
     name: "Qwen/Qwen3.5-397B-A17B",
+    provider: "together",
     reasoning: true,
     compat: {
       supportsReasoningEffort: true,
@@ -344,6 +343,7 @@ export const SYNTHETIC_MODELS: SyntheticModelConfig[] = [
   {
     id: "hf:MiniMaxAI/MiniMax-M2.5",
     name: "MiniMaxAI/MiniMax-M2.5",
+    provider: "synthetic",
     reasoning: true,
     input: ["text"],
     cost: {
@@ -364,6 +364,7 @@ export const SYNTHETIC_MODELS: SyntheticModelConfig[] = [
   {
     id: "hf:nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4",
     name: "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4",
+    provider: "synthetic",
     reasoning: true,
     compat: {
       supportsReasoningEffort: true,

package/src/extensions/web-search/tool.ts

@@ -44,6 +44,12 @@ export function registerSyntheticWebSearchTool(pi: ExtensionAPI): void {
     label: "Synthetic: Web Search",
     description:
       "Search the web using Synthetic's zero-data-retention API. Returns search results with titles, URLs, content snippets, and publication dates. Use for finding documentation, articles, recent information, or any web content. Results are fresh and not cached by Synthetic.",
+    promptSnippet: "Search the web using Synthetic's zero-data-retention API",
+    promptGuidelines: [
+      "Use synthetic_web_search for finding documentation, articles, recent information, or any web content.",
+      "Write specific queries with names, dates, versions, or locations for synthetic_web_search.",
+      "synthetic_web_search results are fresh and not cached by Synthetic.",
+    ],
     parameters: SearchParams,
 
     async execute(