[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_164":165,"_166":167},"routes/_app.x.$handle.$name",{"_95":94,"_32":45,"_103":85,"_27":104,"_13":14,"_105":-5,"_11":106,"_107":108,"_109":110,"_111":112,"_113":114,"_20":-5,"_115":-5,"_116":-5,"_117":25,"_118":119,"_120":49,"_121":-5,"_122":-5,"_123":124},"routes/_app.x.$handle.$name.branch.$branch.version.$version.diff",{"_11":12,"_13":14,"_15":16,"_17":-5,"_18":-5},"project",{"_43":44,"_32":45,"_38":46,"_47":-5,"_48":49,"_50":51,"_52":51,"_53":-5,"_54":-5,"_18":55,"_56":31,"_57":49,"_58":-5,"_59":-5,"_60":61,"_62":63,"_64":-5,"_65":-5,"_66":-5,"_67":-5,"_68":69,"_70":69,"_71":72,"_73":72,"_74":72,"_75":69,"_76":69,"_77":78,"_79":80,"_81":82,"_83":25},"version",244,"changes",[19],"aiChatInfo","userId",{"_20":21,"_22":23,"_24":25,"_26":25,"_27":28,"_29":28},"file",{"_30":31,"_32":28,"_33":34,"_35":37,"_38":39,"_40":14,"_41":42,"_24":25},"previous",{"_30":31,"_32":28,"_33":34,"_35":36},"deleted",false,"deletedHere","path","main.tsx","latestPath","fileId","e5dd47ee-2792-11f0-9d52-569c3dd06744","name","fileType","http","content","import { Hono } from \"npm:hono@3\";\nimport { cors } from \"npm:hono/cors\";\nimport { createOpenAI } from \"npm:@ai-sdk/openai\";\nimport { createAnthropic } from \"npm:@ai-sdk/anthropic@0.0.48\";\nimport { google, createGoogleGenerativeAI } from 'npm:@ai-sdk/google';\nimport { mistral, createMistral } from \"npm:@ai-sdk/mistral\";\nimport { generateText, streamText, generateObject, embed } from \"npm:ai\";\n// import { z } from \"npm:zod\";\n// import { nanoid } from 'npm:nanoid';\nimport stringHash from 'npm:string-hash';\n\nimport { blobby } from \"https://esm.town/v/@yawnxyz/blobby\";\n\nconst URL = \"https://yawnxyz-coversheetai.web.val.run\"\nconst taskPrefix = \"task::ai:\"\n\n\nconst app = new Hono();\n\napp.use('*', cors({\n origin: '*',\n allowMethods: ['GET', 'POST'],\n allowHeaders: ['Content-Type'],\n}));\n\nconst anthropic = createAnthropic({\n // apiKey = Deno.env.get(\"ANTHROPIC_API_KEY\");\n apiKey: Deno.env.get(\"ANTHROPIC_API_KEY_COVERSHEET\")\n});\n\nconst openai = createOpenAI({\n // apiKey = Deno.env.get(\"OPENAI_API_KEY\");\n apiKey: Deno.env.get(\"OPENAI_API_KEY_COVERSHEET\")\n});\n\n\nconst groq = createOpenAI({\n baseURL: 'https://api.groq.com/openai/v1',\n apiKey: Deno.env.get(\"GROQ_API_KEY\"),\n});\n\nconst perplexity = createOpenAI({\n apiKey: Deno.env.get(\"PERPLEXITY_API_KEY\") ?? '',\n baseURL: 'https://api.perplexity.ai/',\n});\n\nconst googleProvider = createGoogleGenerativeAI({\n apiKey: Deno.env.get(\"GOOGLE_GENERATIVE_AI_API_KEY\"),\n});\n\nclass ModelProvider {\n constructor(options = {}) {\n this.id = options.id || 'new-model-provider';\n this.chars = options.chars || {};\n this.system = options.system || '';\n this.memories = options.memories || [];\n this.messages = options.messages || [];\n this.defaultProvider = options.provider || 'openai';\n this.defaultModel = options.model;\n this.defaultMaxTokens = options.maxTokens;\n this.defaultTemperature = options.temperature;\n this.defaultStreaming = options.streaming || false;\n this.defaultSchema = options.schema;\n this.defaultTools = options.tools;\n this.defaultUseMessages = options.useMessages || false;\n this.defaultUseMemory = options.useMemory || false;\n this.defaultUseChars = options.useChars || false;\n this.defaultGetUpdatedChars = options.getUpdatedChars || false;\n this.memoryPrompt = options.memoryPrompt || 'Summarize the key points from the current conversation, focusing on the most important and relevant details for future reference.';\n this.charsPrompt = options.charsPrompt || 'Based on the current conversation, describe how the characters state, emotions, inventory, or ideas might change.';\n }\n\n async gen({\n prompt,\n messages,\n memories,\n chars,\n provider = this.defaultProvider,\n model = this.defaultModel,\n maxTokens = this.defaultMaxTokens,\n temperature = this.defaultTemperature,\n streaming = this.defaultStreaming,\n schema = this.defaultSchema,\n system = this.system,\n tools = this.defaultTools,\n useMessages = this.defaultUseMessages,\n useChars = this.defaultUseChars,\n useMemory = this.defaultUseMemory,\n getUpdatedChars = this.defaultGetUpdatedChars,\n ...additionalSettings\n }) {\n\n if (useMessages) {\n this.useMessages = useMessages;\n if (messages && messages.length > 0) {\n this.messages.push(...messages);\n }\n if (prompt) {\n this.messages.push({ role: 'user', content: prompt });\n prompt = null;\n }\n messages = this.messages;\n }\n\n if (useMemory) {\n this.useMemory = useMemory;\n if (memories && memories.length > 0) {\n this.memories.push(...memories);\n }\n system = system + '\\n\\n' + this.memories.join('\\n\\n');\n }\n\n if (useChars) {\n this.useChars = useChars;\n if (chars) {\n this.chars = { ...this.chars, ...chars };\n }\n system = system + '\\n\\n' + JSON.stringify(this.chars);\n }\n\n console.log('gen():', provider, model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, additionalSettings)\n\n let result;\n switch (provider) {\n case 'openai':\n result = await this.generateOpenAIResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings });\n break;\n case 'anthropic':\n result = await this.generateAnthropicResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings });\n break;\n case 'groq':\n result = await this.generateGroqResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings });\n break;\n case 'perplexity':\n result = await this.generatePerplexityResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings });\n break;\n case 'mistral':\n result = await this.generateMistralResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings });\n break;\n case 'google':\n result = await this.generateGoogleResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings });\n break;\n default:\n throw new Error(`Invalid provider ${provider}`);\n }\n\n const { text, toolResults } = result;\n\n if (this.useMessages) {\n this.messages.push({ role: 'assistant', content: text });\n }\n if (this.useMemory) {\n const memory = await this.generateMemory({ text, messages: this.messages });\n this.memories.push(memory);\n }\n // if (this.useChars && this.getUpdatedChars) {\n // const charsUpdate = await this.updateChars({ messages: this.messages });\n // console.log('Character state updated:', charsUpdate);\n // }\n\n // todo: add logging here\n\n return {\n text,\n toolResults,\n finishReason: result.finishReason,\n usage: result.usage,\n model: result.modelId,\n embedding: result.embedding,\n textStream: result.textStream\n };\n }\n\n async generateOpenAIResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, embed, value, dimensions, user, ...additionalSettings }) {\n const modelId = model || 'gpt-3.5-turbo';\n\n if (embed) {\n let result = await this.generateOpenAIEmbedding({ model, value, dimensions, user });\n // console.log('embed!', result)\n return result\n }\n\n const options = {\n model: openai(modelId),\n system,\n temperature,\n max_tokens: maxTokens || 100,\n ...additionalSettings\n };\n\n if (prompt && messages && messages.length > 0) {\n throw new Error('prompt and messages cannot be defined at the same time');\n } else if (prompt) {\n options.prompt = prompt;\n } else if (messages && messages.length > 0) {\n options.messages = messages;\n }\n\n if (tools) {\n options.tools = tools;\n }\n\n if (schema) {\n const { object } = await generateObject({\n ...options,\n schema,\n });\n return { object, model: modelId };\n } else if (streaming) {\n const { textStream } = await streamText(options);\n return { textStream, model: modelId };\n } else {\n let { text, finishReason, usage, toolResults } = await generateText(options);\n\n if (toolResults && toolResults.length > 0) {\n // we have to do this bc the AI SDK doesn't seem to let you carry over function calling results, \n // so we pretend to be the user instead of using the official role: tool API, which breaks everything\n for (const toolResult of toolResults) {\n options.messages.push({\n role: \"user\",\n content: \": \" + toolResult.result + \"\",\n })\n }\n let res = await generateText(options);\n\n\n // Remove tool result names from options.messages\n this.messages = this.messages.filter(\n message => !toolResults.some(toolResult => message.content.includes(\": \"))\n );\n\n text = res.text;\n }\n return { text, finishReason, usage, toolResults, model: modelId };\n }\n }\n\n async generateOpenAIEmbedding({ model, value, dimensions, user }) {\n const modelId = model || 'text-embedding-3-large';\n const options = {\n model: openai.embedding(modelId, {\n dimensions,\n user,\n }),\n value,\n };\n\n const { embedding } = await embed(options);\n return { embedding, model: modelId };\n }\n\n async generateAnthropicResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings }) {\n const modelId = model || 'claude-3-5-sonnet-20240620';\n\n console.log('generateAnthropicResponse', modelId, `prompt: ${prompt}`, maxTokens, temperature, messages, system)\n\n if (schema) {\n const { object } = await generateObject({\n model: anthropic(modelId),\n system,\n prompt: prompt,\n schema,\n max_tokens: maxTokens || 100,\n temperature,\n messages: messages,\n ...additionalSettings\n });\n return { object, model: modelId };\n } else if (streaming) {\n const { textStream } = await streamText({\n model: anthropic(modelId),\n system,\n prompt: prompt,\n max_tokens: maxTokens || 100,\n messages: messages,\n ...additionalSettings\n });\n return { textStream, model: modelId };\n } else {\n prompt = \"tell me a banana joke\"\n console.log('generateAnthropicResponse', modelId, `prompt: ${prompt}`, maxTokens, temperature, messages, system)\n const { text, finishReason, usage, toolResults } = await generateText({\n model: anthropic(modelId),\n system: system || '',\n prompt: prompt,\n max_tokens: maxTokens || 100,\n messages: messages || null,\n ...additionalSettings\n });\n return { text, finishReason, usage, toolResults, model: modelId };\n }\n }\n\n async generateGroqResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings }) {\n const modelId = model || 'llama3-8b-8192';\n\n if (schema) {\n const { object } = await generateObject({\n model: groq(modelId),\n system:system||'',\n prompt: prompt,\n schema,\n temperature,\n max_tokens: maxTokens || 100,\n messages: messages,\n ...additionalSettings\n });\n return { object, model: modelId };\n } else if (streaming) {\n const { textStream } = await streamText({\n model: groq(modelId),\n system,\n prompt: prompt,\n max_tokens: maxTokens || 100,\n temperature,\n messages: messages,\n ...additionalSettings\n });\n return { textStream, model: modelId };\n } else {\n console.log('generateGroqResponse', modelId, prompt, maxTokens, temperature, messages, system)\n const { text, finishReason, usage, toolResults } = await generateText({\n model: groq(modelId),\n system,\n prompt: prompt,\n max_tokens: maxTokens || 100,\n temperature,\n messages: messages,\n ...additionalSettings\n });\n return { text, finishReason, usage, toolResults, model: modelId };\n }\n }\n\n async generateMistralResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings }) {\n const modelId = model || 'mistral-large-latest';\n\n const options = {\n model: mistral(modelId),\n system,\n max_tokens: maxTokens || 100,\n temperature,\n ...additionalSettings\n };\n\n if (prompt) {\n options.prompt = prompt;\n } else if (messages && messages.length > 0) {\n options.messages = messages;\n }\n\n if (tools) {\n options.tools = tools;\n }\n\n if (schema) {\n const { object } = await generateObject({\n ...options,\n schema,\n });\n return { object, model: modelId };\n } else if (streaming) {\n const { textStream } = await streamText(options);\n return { textStream, model: modelId };\n } else {\n const { text, finishReason, usage, toolResults } = await generateText(options);\n return { text, finishReason, usage, toolResults, model: modelId };\n }\n }\n\n async generatePerplexityResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings }) {\n const modelId = model || 'llama-3-sonar-large-32k-online';\n\n const options = {\n model: perplexity(modelId),\n system,\n max_tokens: maxTokens || 100,\n temperature,\n ...additionalSettings\n };\n\n if (prompt && messages?.length > 0) {\n throw new Error('prompt and messages cannot be defined at the same time');\n } else if (prompt) {\n options.prompt = prompt;\n } else if (messages && messages?.length > 0) {\n options.messages = messages;\n } else {\n throw new Error('Either prompt or messages must be provided');\n }\n\n if (tools) {\n options.tools = tools;\n }\n\n if (schema) {\n const { object } = await generateObject({\n ...options,\n schema,\n });\n return { object };\n } else if (streaming) {\n const { textStream } = await streamText(options);\n return { textStream, model: modelId };\n } else {\n const { text, finishReason, usage, toolResults } = await generateText(options);\n return { text, finishReason, usage, toolResults, model: modelId };\n }\n }\n\n async generateGoogleResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings }) {\n const modelId = model || 'models/gemini-1.5-flash-latest';\n\n const options = {\n model: googleProvider(modelId),\n system,\n max_tokens: maxTokens || 100,\n temperature,\n ...additionalSettings\n };\n\n if (prompt) {\n options.prompt = prompt;\n } else if (messages && messages.length > 0) {\n options.messages = messages;\n }\n\n if (tools) {\n options.tools = tools;\n }\n\n if (schema) {\n const { object } = await generateObject({\n ...options,\n schema,\n });\n return { object, model: modelId };\n } else if (streaming) {\n const { textStream } = await streamText(options);\n return { textStream, model: modelId };\n } else {\n const { text, finishReason, usage, toolResults } = await generateText(options);\n return { text, finishReason, usage, toolResults, model: modelId };\n }\n }\n\n async generateMemory({ provider = this, text, messages, prompt = this.memoryPrompt }) {\n const conversationSummary = messages ? messages.map(msg => `${msg.role}: ${msg.content}`).join('\\n') : '';\n let inputText = `Messages:\\n${conversationSummary}\\n----\\n\\n`;\n if (text) {\n inputText += `\\n\\nAdditional Context: ${text}`;\n }\n\n if (provider) {\n inputText += `Background:\\n-----\\n`;\n inputText += provider.system + '\\n';\n inputText += provider.memories.join('\\n') + '\\n';\n inputText += JSON.stringify(provider.chars) + '\\n';\n inputText += `\\n-----\\n\\n\\n`;\n }\n\n const { text: memory } = await generateText({\n model: anthropic(this.memoryModel || 'claude-3-5-sonnet-20240620'),\n prompt: `${prompt}\\n\\n${inputText}`,\n });\n\n return memory;\n }\n\n async updateChars({ provider = this, text, messages, prompt = this.charsPrompt }) {\n const conversationSummary = messages ? messages.map(msg => `${msg.role}: ${msg.content}`).join('\\n') : '';\n let inputText = `Messages:\\n${conversationSummary}\\n----\\n\\n`;\n if (text) {\n inputText += `\\n\\nAdditional Context: ${text}`;\n }\n if (provider) {\n inputText += `Background:\\n-----\\n`;\n inputText += provider.system + '\\n';\n inputText += provider.memories.join('\\n') + '\\n';\n inputText += \"\" + JSON.stringify(provider.chars) + '\\n';\n inputText += `\\n-----\\n\\n\\n`;\n }\n\n const { text: charsUpdate } = await generateText({\n model: anthropic(this.charsModel || 'claude-3-5-sonnet-20240620'),\n prompt: `${prompt}\\n\\n${inputText}`,\n });\n\n const updatedChars = JSON.parse(charsUpdate);\n this.chars = { ...this.chars, ...updatedChars };\n return charsUpdate;\n }\n}\n\nconst modelProvider = new ModelProvider();\nconst gen = modelProvider.gen;\n\napp.get('/generate', async (c) => {\n const provider = c.req.query('provider') || 'openai';\n const model = c.req.query('model');\n const prompt = c.req.query('prompt');\n const maxTokens = parseInt(c.req.query('max_tokens')) || undefined;\n const temperature = parseFloat(c.req.query('temperature')) || undefined;\n const streaming = c.req.query('streaming') === 'true';\n const schema = JSON.parse(c.req.query('schema') || 'null');\n const system = c.req.query('system');\n const messages = JSON.parse(c.req.query('messages') || 'null');\n const tools = JSON.parse(c.req.query('tools') || 'null');\n const embed = JSON.parse(c.req.query('embed') || 'null');\n const value = c.req.query('value');\n const dimensions = c.req.query('dimensions') ? Number(c.req.query('dimensions')) : null;\n\n\n try {\n console.log(\"get/generate:\", { mode: 'get/generate', prompt, provider, model });\n const response = await modelProvider.gen({ provider, model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, embed, value });\n console.log('response:', response)\n return c.json(response);\n } catch (error) {\n console.error('API error:', error);\n return c.text('Error occurred.', 500);\n }\n});\n\napp.post('/generate', async (c) => {\n const { provider, model, prompt, max_tokens, temperature, streaming, schema, system, messages, tools, embed, value, dimensions, user, ...additionalSettings } = await c.req.json();\n\n try {\n console.log(\"post/generate\", { mode: 'post/generate', prompt, provider, model });\n const response = await modelProvider.gen({\n provider: provider || 'openai',\n model,\n prompt,\n maxTokens: max_tokens,\n temperature,\n streaming: streaming || false,\n schema,\n system,\n messages,\n tools,\n embed,\n value,\n dimensions,\n user,\n ...additionalSettings\n });\n console.log(response);\n return c.json(response);\n } catch (error) {\n console.error('API error:', error);\n return c.text('Error occurred.', 500);\n }\n});\n\n\n\n\n\n// this launches a task based on the setting / prompt, and returns the empty task object\n// this prevents the constant google sheets regeneration abuse\n// also lets long-running tasks to complete\napp.post('/task', async (c) => {\n\n let { taskId, useCache = true, ...additionalSettings } = await c.req.json();\n\n\n // taskid is based on a simple stringhash of the settings, which doubles as a way to check if the task has already been done\n // this catches lots of duplicate requests from Google Sheets\n taskId = taskId || `${taskPrefix}${stringHash(JSON.stringify(additionalSettings))}`;\n\n\n let taskObj = await blobby.get(taskId);\n console.log('>>>> /task:', taskId, taskObj)\n if ((taskObj && taskObj.output && useCache) || (taskObj && taskObj.error)) {\n return c.json(taskObj);\n }\n\n taskObj = {\n taskId,\n dateAdded: new Date().toISOString(),\n input: additionalSettings,\n output: null\n }\n\n // just launch it, don't wait for the result\n // const taskRunResponse = await fetch(`${URL}/taskrun`, {\n const taskRunResponse = fetch(`${URL}/taskrun`, {\n method: 'POST',\n headers: {\n 'Content-Type': 'application/json'\n },\n body: JSON.stringify({ ...taskObj, ...additionalSettings })\n });\n\n console.log('[task] task launched!, taskObj:', taskObj)\n\n // return the empty task object\n return c.json(taskObj);\n});\n\n\n\napp.post('/taskrun', async (c) => {\n let { taskId, ...settings } = await c.req.json();\n\n // Generate a unique taskId if not provided\n // DON'T DO THIS!!! This adds more settings so creates a new task lol\n // but thankfully the taskId should always be passed in\n // taskId = taskId || `${taskPrefix}${stringHash(JSON.stringify(settings))}`;\n if(!taskId) return c.error('No taskId provided', 400);\n\n let taskObj = await blobby.get(taskId);\n console.log('>>>> /taskrun:', taskId, taskObj)\n if (taskObj && taskObj.output) {\n console.log('returning cached task:', { taskId, settings })\n return c.json(taskObj);\n }\n\n\n console.log('running tasks with:', { taskId, settings })\n\n try {\n // Run the modelProvider with the provided settings\n const output = await modelProvider.gen(settings);\n\n // Retrieve the task object from blobby\n let taskObj = await blobby.get(taskId);\n\n // Update the task object with the output\n taskObj = {\n taskId,\n ...taskObj,\n ...settings,\n output,\n dateCompleted: new Date().toISOString()\n };\n\n // Store the updated task object in blobby\n await blobby.set(taskId, taskObj);\n\n console.log('[taskrun] Task run complete:', taskObj);\n return c.json(taskObj);\n } catch (error) {\n console.error('Task run error:', error);\n\n // Update the task object with the error\n taskObj = {\n taskId,\n ...taskObj,\n ...settings,\n dateCompleted: new Date().toISOString(),\n error: error,\n };\n\n // Store the updated task object in blobby\n await blobby.set(taskId, taskObj);\n\n let errorObj = await blobby.get(taskId);\n console.error('Errored task object:', errorObj);\n\n\n return c.text('Error occurred during task run.', 500);\n }\n});\n\n\n// very simple wrapper, mostly for testing / making sure things work\nconst ai = async (input, opts={}) => {\n if (typeof input === 'string') {\n console.log('ai input:', input, opts)\n let result = await modelProvider.gen({ prompt: input, ...opts });\n console.log(\"[ai]:\" + result.text);\n return result.text;\n } else {\n return modelProvider.gen(input);\n }\n}\n\nfunction test() {\n console.log(\"test!\")\n}\n\nexport default app.fetch;\nexport { ModelProvider, modelProvider, ai, test };\n","import { Hono } from \"npm:hono@3\";\nimport { cors } from \"npm:hono/cors\";\nimport { createOpenAI } from \"npm:@ai-sdk/openai\";\nimport { createAnthropic } from \"npm:@ai-sdk/anthropic@0.0.48\";\nimport { google, createGoogleGenerativeAI } from 'npm:@ai-sdk/google';\nimport { mistral, createMistral } from \"npm:@ai-sdk/mistral\";\nimport { generateText, streamText, generateObject, embed } from \"npm:ai\";\n// import { z } from \"npm:zod\";\n// import { nanoid } from 'npm:nanoid';\nimport stringHash from 'npm:string-hash';\n\nimport { blobby } from \"https://esm.town/v/@yawnxyz/blobby\";\n\nconst URL = \"https://yawnxyz-coversheetai.web.val.run\"\nconst taskPrefix = \"task::ai:\"\n\n\nconst app = new Hono();\n\napp.use('*', cors({\n origin: '*',\n allowMethods: ['GET', 'POST'],\n allowHeaders: ['Content-Type'],\n}));\n\nconst anthropic = createAnthropic({\n // apiKey = Deno.env.get(\"ANTHROPIC_API_KEY\");\n apiKey: Deno.env.get(\"ANTHROPIC_API_KEY_COVERSHEET\")\n});\n\nconst openai = createOpenAI({\n // apiKey = Deno.env.get(\"OPENAI_API_KEY\");\n apiKey: Deno.env.get(\"OPENAI_API_KEY_COVERSHEET\")\n});\n\n\nconst groq = createOpenAI({\n baseURL: 'https://api.groq.com/openai/v1',\n apiKey: Deno.env.get(\"GROQ_API_KEY\"),\n});\n\nconst perplexity = createOpenAI({\n apiKey: Deno.env.get(\"PERPLEXITY_API_KEY\") ?? '',\n baseURL: 'https://api.perplexity.ai/',\n});\n\nconst googleProvider = createGoogleGenerativeAI({\n apiKey: Deno.env.get(\"GOOGLE_GENERATIVE_AI_API_KEY\"),\n});\n\nclass ModelProvider {\n constructor(options = {}) {\n this.id = options.id || 'new-model-provider';\n this.chars = options.chars || {};\n this.system = options.system || '';\n this.memories = options.memories || [];\n this.messages = options.messages || [];\n this.defaultProvider = options.provider || 'openai';\n this.defaultModel = options.model;\n this.defaultMaxTokens = options.maxTokens;\n this.defaultTemperature = options.temperature;\n this.defaultStreaming = options.streaming || false;\n this.defaultSchema = options.schema;\n this.defaultTools = options.tools;\n this.defaultUseMessages = options.useMessages || false;\n this.defaultUseMemory = options.useMemory || false;\n this.defaultUseChars = options.useChars || false;\n this.defaultGetUpdatedChars = options.getUpdatedChars || false;\n this.memoryPrompt = options.memoryPrompt || 'Summarize the key points from the current conversation, focusing on the most important and relevant details for future reference.';\n this.charsPrompt = options.charsPrompt || 'Based on the current conversation, describe how the characters state, emotions, inventory, or ideas might change.';\n }\n\n async gen({\n prompt,\n messages,\n memories,\n chars,\n provider = this.defaultProvider,\n model = this.defaultModel,\n maxTokens = this.defaultMaxTokens,\n temperature = this.defaultTemperature,\n streaming = this.defaultStreaming,\n schema = this.defaultSchema,\n system = this.system,\n tools = this.defaultTools,\n useMessages = this.defaultUseMessages,\n useChars = this.defaultUseChars,\n useMemory = this.defaultUseMemory,\n getUpdatedChars = this.defaultGetUpdatedChars,\n ...additionalSettings\n }) {\n\n if (useMessages) {\n this.useMessages = useMessages;\n if (messages && messages.length > 0) {\n this.messages.push(...messages);\n }\n if (prompt) {\n this.messages.push({ role: 'user', content: prompt });\n prompt = null;\n }\n messages = this.messages;\n }\n\n if (useMemory) {\n this.useMemory = useMemory;\n if (memories && memories.length > 0) {\n this.memories.push(...memories);\n }\n system = system + '\\n\\n' + this.memories.join('\\n\\n');\n }\n\n if (useChars) {\n this.useChars = useChars;\n if (chars) {\n this.chars = { ...this.chars, ...chars };\n }\n system = system + '\\n\\n' + JSON.stringify(this.chars);\n }\n\n console.log('gen():', provider, model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, additionalSettings)\n\n let result;\n switch (provider) {\n case 'openai':\n result = await this.generateOpenAIResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings });\n break;\n case 'anthropic':\n result = await this.generateAnthropicResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings });\n break;\n case 'groq':\n result = await this.generateGroqResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings });\n break;\n case 'perplexity':\n result = await this.generatePerplexityResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings });\n break;\n case 'mistral':\n result = await this.generateMistralResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings });\n break;\n case 'google':\n result = await this.generateGoogleResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings });\n break;\n default:\n throw new Error(`Invalid provider ${provider}`);\n }\n\n const { text, toolResults } = result;\n\n if (this.useMessages) {\n this.messages.push({ role: 'assistant', content: text });\n }\n if (this.useMemory) {\n const memory = await this.generateMemory({ text, messages: this.messages });\n this.memories.push(memory);\n }\n // if (this.useChars && this.getUpdatedChars) {\n // const charsUpdate = await this.updateChars({ messages: this.messages });\n // console.log('Character state updated:', charsUpdate);\n // }\n\n // todo: add logging here\n\n return {\n text,\n toolResults,\n finishReason: result.finishReason,\n usage: result.usage,\n model: result.modelId,\n embedding: result.embedding,\n textStream: result.textStream\n };\n }\n\n async generateOpenAIResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, embed, value, dimensions, user, ...additionalSettings }) {\n const modelId = model || 'gpt-3.5-turbo';\n\n if (embed) {\n let result = await this.generateOpenAIEmbedding({ model, value, dimensions, user });\n // console.log('embed!', result)\n return result\n }\n\n const options = {\n model: openai(modelId),\n system,\n temperature,\n max_tokens: maxTokens || 100,\n ...additionalSettings\n };\n\n if (prompt && messages && messages.length > 0) {\n throw new Error('prompt and messages cannot be defined at the same time');\n } else if (prompt) {\n options.prompt = prompt;\n } else if (messages && messages.length > 0) {\n options.messages = messages;\n }\n\n if (tools) {\n options.tools = tools;\n }\n\n if (schema) {\n const { object } = await generateObject({\n ...options,\n schema,\n });\n return { object, model: modelId };\n } else if (streaming) {\n const { textStream } = await streamText(options);\n return { textStream, model: modelId };\n } else {\n let { text, finishReason, usage, toolResults } = await generateText(options);\n\n if (toolResults && toolResults.length > 0) {\n // we have to do this bc the AI SDK doesn't seem to let you carry over function calling results, \n // so we pretend to be the user instead of using the official role: tool API, which breaks everything\n for (const toolResult of toolResults) {\n options.messages.push({\n role: \"user\",\n content: \": \" + toolResult.result + \"\",\n })\n }\n let res = await generateText(options);\n\n\n // Remove tool result names from options.messages\n this.messages = this.messages.filter(\n message => !toolResults.some(toolResult => message.content.includes(\": \"))\n );\n\n text = res.text;\n }\n return { text, finishReason, usage, toolResults, model: modelId };\n }\n }\n\n async generateOpenAIEmbedding({ model, value, dimensions, user }) {\n const modelId = model || 'text-embedding-3-large';\n const options = {\n model: openai.embedding(modelId, {\n dimensions,\n user,\n }),\n value,\n };\n\n const { embedding } = await embed(options);\n return { embedding, model: modelId };\n }\n\n async generateAnthropicResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings }) {\n const modelId = model || 'claude-3-5-sonnet-20240620';\n\n console.log('generateAnthropicResponse', modelId, `prompt: ${prompt}`, maxTokens, temperature, messages, system)\n\n if (schema) {\n const { object } = await generateObject({\n model: anthropic(modelId),\n system,\n prompt: prompt,\n schema,\n max_tokens: maxTokens || 100,\n temperature,\n messages: messages,\n ...additionalSettings\n });\n return { object, model: modelId };\n } else if (streaming) {\n const { textStream } = await streamText({\n model: anthropic(modelId),\n system,\n prompt: prompt,\n max_tokens: maxTokens || 100,\n messages: messages,\n ...additionalSettings\n });\n return { textStream, model: modelId };\n } else {\n prompt = \"tell me a banana joke\"\n console.log('generateAnthropicResponse/text', modelId, `prompt: ${prompt}`, maxTokens, temperature, messages, system, additionalSettings)\n const { text, finishReason, usage, toolResults } = await generateText({\n model: anthropic(modelId),\n system: system || '',\n prompt: prompt,\n max_tokens: maxTokens || 100,\n messages: messages || null,\n ...additionalSettings\n });\n return { text, finishReason, usage, toolResults, model: modelId };\n }\n }\n\n async generateGroqResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings }) {\n const modelId = model || 'llama3-8b-8192';\n\n if (schema) {\n const { object } = await generateObject({\n model: groq(modelId),\n system:system||'',\n prompt: prompt,\n schema,\n temperature,\n max_tokens: maxTokens || 100,\n messages: messages,\n ...additionalSettings\n });\n return { object, model: modelId };\n } else if (streaming) {\n const { textStream } = await streamText({\n model: groq(modelId),\n system,\n prompt: prompt,\n max_tokens: maxTokens || 100,\n temperature,\n messages: messages,\n ...additionalSettings\n });\n return { textStream, model: modelId };\n } else {\n console.log('generateGroqResponse', modelId, prompt, maxTokens, temperature, messages, system)\n const { text, finishReason, usage, toolResults } = await generateText({\n model: groq(modelId),\n system,\n prompt: prompt,\n max_tokens: maxTokens || 100,\n temperature,\n messages: messages,\n ...additionalSettings\n });\n return { text, finishReason, usage, toolResults, model: modelId };\n }\n }\n\n async generateMistralResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings }) {\n const modelId = model || 'mistral-large-latest';\n\n const options = {\n model: mistral(modelId),\n system,\n max_tokens: maxTokens || 100,\n temperature,\n ...additionalSettings\n };\n\n if (prompt) {\n options.prompt = prompt;\n } else if (messages && messages.length > 0) {\n options.messages = messages;\n }\n\n if (tools) {\n options.tools = tools;\n }\n\n if (schema) {\n const { object } = await generateObject({\n ...options,\n schema,\n });\n return { object, model: modelId };\n } else if (streaming) {\n const { textStream } = await streamText(options);\n return { textStream, model: modelId };\n } else {\n const { text, finishReason, usage, toolResults } = await generateText(options);\n return { text, finishReason, usage, toolResults, model: modelId };\n }\n }\n\n async generatePerplexityResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings }) {\n const modelId = model || 'llama-3-sonar-large-32k-online';\n\n const options = {\n model: perplexity(modelId),\n system,\n max_tokens: maxTokens || 100,\n temperature,\n ...additionalSettings\n };\n\n if (prompt && messages?.length > 0) {\n throw new Error('prompt and messages cannot be defined at the same time');\n } else if (prompt) {\n options.prompt = prompt;\n } else if (messages && messages?.length > 0) {\n options.messages = messages;\n } else {\n throw new Error('Either prompt or messages must be provided');\n }\n\n if (tools) {\n options.tools = tools;\n }\n\n if (schema) {\n const { object } = await generateObject({\n ...options,\n schema,\n });\n return { object };\n } else if (streaming) {\n const { textStream } = await streamText(options);\n return { textStream, model: modelId };\n } else {\n const { text, finishReason, usage, toolResults } = await generateText(options);\n return { text, finishReason, usage, toolResults, model: modelId };\n }\n }\n\n async generateGoogleResponse({ model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, ...additionalSettings }) {\n const modelId = model || 'models/gemini-1.5-flash-latest';\n\n const options = {\n model: googleProvider(modelId),\n system,\n max_tokens: maxTokens || 100,\n temperature,\n ...additionalSettings\n };\n\n if (prompt) {\n options.prompt = prompt;\n } else if (messages && messages.length > 0) {\n options.messages = messages;\n }\n\n if (tools) {\n options.tools = tools;\n }\n\n if (schema) {\n const { object } = await generateObject({\n ...options,\n schema,\n });\n return { object, model: modelId };\n } else if (streaming) {\n const { textStream } = await streamText(options);\n return { textStream, model: modelId };\n } else {\n const { text, finishReason, usage, toolResults } = await generateText(options);\n return { text, finishReason, usage, toolResults, model: modelId };\n }\n }\n\n async generateMemory({ provider = this, text, messages, prompt = this.memoryPrompt }) {\n const conversationSummary = messages ? messages.map(msg => `${msg.role}: ${msg.content}`).join('\\n') : '';\n let inputText = `Messages:\\n${conversationSummary}\\n----\\n\\n`;\n if (text) {\n inputText += `\\n\\nAdditional Context: ${text}`;\n }\n\n if (provider) {\n inputText += `Background:\\n-----\\n`;\n inputText += provider.system + '\\n';\n inputText += provider.memories.join('\\n') + '\\n';\n inputText += JSON.stringify(provider.chars) + '\\n';\n inputText += `\\n-----\\n\\n\\n`;\n }\n\n const { text: memory } = await generateText({\n model: anthropic(this.memoryModel || 'claude-3-5-sonnet-20240620'),\n prompt: `${prompt}\\n\\n${inputText}`,\n });\n\n return memory;\n }\n\n async updateChars({ provider = this, text, messages, prompt = this.charsPrompt }) {\n const conversationSummary = messages ? messages.map(msg => `${msg.role}: ${msg.content}`).join('\\n') : '';\n let inputText = `Messages:\\n${conversationSummary}\\n----\\n\\n`;\n if (text) {\n inputText += `\\n\\nAdditional Context: ${text}`;\n }\n if (provider) {\n inputText += `Background:\\n-----\\n`;\n inputText += provider.system + '\\n';\n inputText += provider.memories.join('\\n') + '\\n';\n inputText += \"\" + JSON.stringify(provider.chars) + '\\n';\n inputText += `\\n-----\\n\\n\\n`;\n }\n\n const { text: charsUpdate } = await generateText({\n model: anthropic(this.charsModel || 'claude-3-5-sonnet-20240620'),\n prompt: `${prompt}\\n\\n${inputText}`,\n });\n\n const updatedChars = JSON.parse(charsUpdate);\n this.chars = { ...this.chars, ...updatedChars };\n return charsUpdate;\n }\n}\n\nconst modelProvider = new ModelProvider();\nconst gen = modelProvider.gen;\n\napp.get('/generate', async (c) => {\n const provider = c.req.query('provider') || 'openai';\n const model = c.req.query('model');\n const prompt = c.req.query('prompt');\n const maxTokens = parseInt(c.req.query('max_tokens')) || undefined;\n const temperature = parseFloat(c.req.query('temperature')) || undefined;\n const streaming = c.req.query('streaming') === 'true';\n const schema = JSON.parse(c.req.query('schema') || 'null');\n const system = c.req.query('system');\n const messages = JSON.parse(c.req.query('messages') || 'null');\n const tools = JSON.parse(c.req.query('tools') || 'null');\n const embed = JSON.parse(c.req.query('embed') || 'null');\n const value = c.req.query('value');\n const dimensions = c.req.query('dimensions') ? Number(c.req.query('dimensions')) : null;\n\n\n try {\n console.log(\"get/generate:\", { mode: 'get/generate', prompt, provider, model });\n const response = await modelProvider.gen({ provider, model, prompt, maxTokens, temperature, streaming, schema, system, messages, tools, embed, value });\n console.log('response:', response)\n return c.json(response);\n } catch (error) {\n console.error('API error:', error);\n return c.text('Error occurred.', 500);\n }\n});\n\napp.post('/generate', async (c) => {\n const { provider, model, prompt, max_tokens, temperature, streaming, schema, system, messages, tools, embed, value, dimensions, user, ...additionalSettings } = await c.req.json();\n\n try {\n console.log(\"post/generate\", { mode: 'post/generate', prompt, provider, model });\n const response = await modelProvider.gen({\n provider: provider || 'openai',\n model,\n prompt,\n maxTokens: max_tokens,\n temperature,\n streaming: streaming || false,\n schema,\n system,\n messages,\n tools,\n embed,\n value,\n dimensions,\n user,\n ...additionalSettings\n });\n console.log(response);\n return c.json(response);\n } catch (error) {\n console.error('API error:', error);\n return c.text('Error occurred.', 500);\n }\n});\n\n\n\n\n\n// this launches a task based on the setting / prompt, and returns the empty task object\n// this prevents the constant google sheets regeneration abuse\n// also lets long-running tasks to complete\napp.post('/task', async (c) => {\n\n let { taskId, useCache = true, ...additionalSettings } = await c.req.json();\n\n\n // taskid is based on a simple stringhash of the settings, which doubles as a way to check if the task has already been done\n // this catches lots of duplicate requests from Google Sheets\n taskId = taskId || `${taskPrefix}${stringHash(JSON.stringify(additionalSettings))}`;\n\n\n let taskObj = await blobby.get(taskId);\n console.log('>>>> /task:', taskId, taskObj)\n if ((taskObj && taskObj.output && useCache) || (taskObj && taskObj.error)) {\n return c.json(taskObj);\n }\n\n taskObj = {\n taskId,\n dateAdded: new Date().toISOString(),\n input: additionalSettings,\n output: null\n }\n\n // just launch it, don't wait for the result\n // const taskRunResponse = await fetch(`${URL}/taskrun`, {\n const taskRunResponse = fetch(`${URL}/taskrun`, {\n method: 'POST',\n headers: {\n 'Content-Type': 'application/json'\n },\n body: JSON.stringify({ ...taskObj, ...additionalSettings })\n });\n\n console.log('[task] task launched!, taskObj:', taskObj)\n\n // return the empty task object\n return c.json(taskObj);\n});\n\n\n\napp.post('/taskrun', async (c) => {\n let { taskId, ...settings } = await c.req.json();\n\n // Generate a unique taskId if not provided\n // DON'T DO THIS!!! This adds more settings so creates a new task lol\n // but thankfully the taskId should always be passed in\n // taskId = taskId || `${taskPrefix}${stringHash(JSON.stringify(settings))}`;\n if(!taskId) return c.error('No taskId provided', 400);\n\n let taskObj = await blobby.get(taskId);\n console.log('>>>> /taskrun:', taskId, taskObj)\n if (taskObj && taskObj.output) {\n console.log('returning cached task:', { taskId, settings })\n return c.json(taskObj);\n }\n\n\n console.log('running tasks with:', { taskId, settings })\n\n try {\n // Run the modelProvider with the provided settings\n const output = await modelProvider.gen(settings);\n\n // Retrieve the task object from blobby\n let taskObj = await blobby.get(taskId);\n\n // Update the task object with the output\n taskObj = {\n taskId,\n ...taskObj,\n ...settings,\n output,\n dateCompleted: new Date().toISOString()\n };\n\n // Store the updated task object in blobby\n await blobby.set(taskId, taskObj);\n\n console.log('[taskrun] Task run complete:', taskObj);\n return c.json(taskObj);\n } catch (error) {\n console.error('Task run error:', error);\n\n // Update the task object with the error\n taskObj = {\n taskId,\n ...taskObj,\n ...settings,\n dateCompleted: new Date().toISOString(),\n error: error,\n };\n\n // Store the updated task object in blobby\n await blobby.set(taskId, taskObj);\n\n let errorObj = await blobby.get(taskId);\n console.error('Errored task object:', errorObj);\n\n\n return c.text('Error occurred during task run.', 500);\n }\n});\n\n\n// very simple wrapper, mostly for testing / making sure things work\nconst ai = async (input, opts={}) => {\n if (typeof input === 'string') {\n console.log('ai input:', input, opts)\n let result = await modelProvider.gen({ prompt: input, ...opts });\n console.log(\"[ai]:\" + result.text);\n return result.text;\n } else {\n return modelProvider.gen(input);\n }\n}\n\nfunction test() {\n console.log(\"test!\")\n}\n\nexport default app.fetch;\nexport { ModelProvider, modelProvider, ai, test };\n","createdAt",["D",1724969892350],"branchVersion","branchId","e5dad7ac-2792-11f0-9d52-569c3dd06744","id","e5da88ba-2792-11f0-9d52-569c3dd06744","ai",["D",1746217232347],"deletedAt","legacyEnvVars",true,"privacy","public","httpPrivacy","description","imageUrl","a8270e74-6479-475a-8025-9b69544ee0e7","pinnedFileId","pinPreview","pinMigrated","screenshotVersion","likeCount",2,"orgId","8319defa-6e1a-46d4-ba66-2114ffd3360b","tursoName","tursoHostname","tursoMeasuredAt","pinned","userPerm","r","valPerm","blobPerm","rw","emailPerm","sqlitePerm","projectPerm","telemetryPerm","user",{"_95":94,"_102":98},"org",{"_95":94,"_96":-5,"_97":98,"_18":55,"_99":100,"_101":25},"branch",{"_13":84,"_32":85,"_43":42,"_86":44,"_87":-5,"_88":-5,"_89":-5,"_90":-5,"_91":-5,"_92":45,"_93":94},"isOwned",252,"main","projectId","forkedBranchId","forkedBranchVersion","forkedProjectPrivacy","forkedProjectId","forkedProjectOrgId","projectName","userHandle","yawnxyz","handle","displayName","avatar","https://images.clerk.dev/oauth_github/img_2NnaHhpxNuH1xWRIRjQNoo16TVc.jpeg","tier","pro","canRestrictVals","profileImageUrl","branchName","","adoptOrgId",{"_43":44,"_32":45,"_38":160,"_47":-5,"_48":49,"_50":51,"_52":51,"_53":-5,"_54":-5,"_18":55,"_56":31,"_57":49,"_58":-5,"_59":-5,"_60":61,"_62":63,"_64":-5,"_65":-5,"_66":-5,"_67":-5,"_68":69,"_70":69,"_71":72,"_73":72,"_74":72,"_75":69,"_76":69,"_77":161,"_79":162,"_81":163,"_83":25},"tree",[151,152],"files",[151,152],"counts",{"_141":139,"_142":139,"_143":139,"_144":145,"_146":147},"contents",[125,126],"upstreamProject","upstreamBranch","emptyEnvVars","flags",[],"hasHttpFiles","grants","orgs","customDomains",{},{"_43":134,"_32":135,"_30":136,"_128":-5,"_41":42,"_40":137,"_129":49,"_38":138,"_13":139,"_33":20,"_35":-5,"_24":25,"_132":-5,"_27":140},{"_43":127,"_32":28,"_30":31,"_128":-5,"_41":42,"_40":14,"_129":25,"_38":130,"_13":131,"_33":34,"_35":-5,"_24":25,"_132":-5,"_27":133},"ec046094-2792-11f0-9d52-569c3dd06744","parentId","latest",["D",1724969892350],243,"label",[28],"e5dd62a6-2792-11f0-9d52-569c3dd06744","README.md","e5dd6328-2792-11f0-9d52-569c3dd06744",1,["D",1715136553349],0,[135],"pulls","branches","forks","envVars",{"_148":139,"_150":139},"connections",{"_148":139,"_149":139},"count","unfilledCount","emptyCount",{"_43":134,"_32":135,"_30":136,"_128":-5,"_41":42,"_40":137,"_129":49,"_38":157,"_13":139,"_33":20,"_35":-5,"_24":25,"_132":-5,"_154":158,"_27":159},{"_43":127,"_32":28,"_30":31,"_128":-5,"_41":42,"_40":14,"_129":25,"_38":153,"_13":131,"_33":34,"_35":-5,"_24":25,"_132":-5,"_154":155,"_27":156},["D",1724969892350],"children",[],[28],["D",1715136553349],[],[135],["D",1746217232347],{"_95":94,"_102":98},{"_95":94,"_96":-5,"_97":98,"_18":55,"_99":100,"_101":25},{"_13":84,"_32":85,"_43":42,"_86":44,"_87":-5,"_88":-5,"_89":-5,"_90":-5,"_91":-5,"_92":45,"_93":94},"ctx",{"_77":-5,"_197":198,"_122":199,"_200":201,"_202":203,"_204":25,"_205":25,"_206":25,"_207":49,"_208":-5,"_209":-7,"_210":-7,"_211":139,"_212":25,"_213":25,"_214":-5},"ENV",{"_168":169,"_170":171,"_172":85,"_173":174,"_175":176,"_177":178,"_179":180,"_181":182,"_183":184,"_185":186,"_187":188,"_189":190,"_191":192,"_193":194,"_195":196},"REACT_APP_SENTRY_DSN","https://586020d86f05489db63a554690acf1e9@o1398359.ingest.sentry.io/6724499","RENDER_GIT_COMMIT","fdc99cef6780d301d68708a0b7e96c8be98fdd58","RENDER_GIT_BRANCH","POSTHOG_API_TOKEN","phc_nmoyuAkStEZuUp69rYlg1DZgrIt19Gx0Y4Hwf8tGgJP","POSTHOG_API_HOST","https://wart.val.town","WEB_API_TEMPLATE","https://{val}.web.val.run","APP_URL","https://www.val.town","MODULE_URL","https://esm.town","VANITY_WEB_API_TEMPLATE","https://{label}.val.run/","DENOLS_URL","https://lsp.val.town","NODE_ENV","production","CLOUDFLARE_TURNSTILE_SITE_KEY","0x4AAAAAACtaIZJH7s9rBMZ6","GOOGLE_CLIENT_ID","875989696899-jtsqkhlbdv5jrlsk8631p7ipjg3c6o62.apps.googleusercontent.com","STRIPE_PUBLISHABLE_KEY","pk_live_51N2zvWEpZTTLipIkVEwXrUG9AT0yeOOEYxszarZT1jpsZV58Ly6wlg3SltZpHN26CUalOBmUSY7adQRQiSkCZz2500qSXNkf8Y","API_URL","https://api.val.town","aiChatModels",[215,216,217,218,219],[],"preferencesKeybindings","default","preferencesTheme","light","preferencesFixers","preferencesCopilot","preferencesSelectionAi","preferencesCloseBrackets","preferencesChatPanelState","preferencesDprint","preferencesDenofmt","notificationCount","flagDenols","navCollapsed","currentOrgId",{"_220":244,"_132":245,"_223":246,"_53":247,"_226":49,"_227":49,"_228":229,"_230":49},{"_220":239,"_132":240,"_223":241,"_53":242,"_226":49,"_227":25,"_228":243,"_230":25},{"_220":235,"_132":236,"_223":237,"_53":238,"_226":49,"_227":49,"_228":229,"_230":25},{"_220":231,"_132":232,"_223":233,"_53":234,"_226":49,"_227":49,"_228":229,"_230":25},{"_220":221,"_132":222,"_223":224,"_53":225,"_226":49,"_227":49,"_228":229,"_230":25},"value","fable","Claude Fable 5.1","shortLabel","Fable","Most capable; 2× Opus pricing","imageInput","thinking","contextWindowSize",1000000,"isDefault","opus55","Claude Opus 5.5","Opus","Highly capable for complex work","sonnet55","Claude Sonnet 5.5","Sonnet","Balanced speed and capability","haiku","Claude Haiku 4.5","Haiku","Fastest; best for simple edits",200000,"deepseekflash","DeepSeek V4.1 Flash","DeepSeek Flash","Smart and very cheap"]

Diff · v244