Merge pull request #32 from sgomez/release-1.0

Prepare release 1.0

.github/workflows/ci.yml

@@ -19,8 +19,6 @@ jobs:
 
       - name: Setup pnpm
         uses: pnpm/[email protected]
-        with:
-          version: 9.6.0
 
       - name: Use Node.js ${{ matrix.node-version }}
         uses: actions/setup-node@v2

.github/workflows/quality.yml

@@ -16,8 +16,6 @@ jobs:
 
       - name: Setup pnpm
         uses: pnpm/[email protected]
-        with:
-          version: 9.6.0
 
       - name: Use Node.js 18
         uses: actions/setup-node@v3
@@ -40,8 +38,6 @@ jobs:
 
       - name: Setup pnpm
         uses: pnpm/[email protected]
-        with:
-          version: 9.6.0
 
       - name: Use Node.js 18
         uses: actions/setup-node@v3
@@ -64,8 +60,6 @@ jobs:
 
       - name: Setup pnpm
         uses: pnpm/[email protected]
-        with:
-          version: 9.6.0
 
       - name: Use Node.js 18
         uses: actions/setup-node@v3

examples/ai-core/package.json

@@ -4,10 +4,10 @@
   "private": true,
   "scripts": {},
   "dependencies": {
-    "@opentelemetry/sdk-node": "0.52.0",
+    "@opentelemetry/sdk-node": "0.54.2",
     "@opentelemetry/auto-instrumentations-node": "0.47.0",
-    "@opentelemetry/sdk-trace-node": "1.25.0",
-    "ai": "latest",
+    "@opentelemetry/sdk-trace-node": "1.27.0",
+    "ai": "^4.0.0",
     "commander": "^12.1.0",
     "mathjs": "12.4.2",
     "ollama-ai-provider": "workspace:latest",
@@ -17,6 +17,6 @@
   "devDependencies": {
     "@types/node": "20.11.20",
     "tsx": "4.7.1",
-    "typescript": "5.5.4"
+    "typescript": "5.6.3"
   }
 }

examples/ai-core/src/generate-object/ollama-multimodal.ts

@@ -33,4 +33,4 @@ async function main(model: Parameters<typeof ollama>[0]) {
   console.log(JSON.stringify(object.artwork, null, 2))
 }
 
-buildProgram('llava:13b', main).catch(console.error)
+buildProgram('llama3.2-vision', main).catch(console.error)

examples/ai-core/src/generate-text/ollama-multi-step-continue.ts

@@ -0,0 +1,26 @@
+#! /usr/bin/env -S pnpm tsx
+
+import { generateText } from 'ai'
+import { ollama } from 'ollama-ai-provider'
+
+import { buildProgram } from '../tools/command'
+
+async function main(model: Parameters<typeof ollama>[0]) {
+  const { steps, text, usage } = await generateText({
+    experimental_continueSteps: true, // 4096 output tokens
+    maxSteps: 5,
+    model: ollama(model),
+    prompt:
+      'Write a book about Roman history, ' +
+      'from the founding of the city of Rome ' +
+      'to the fall of the Western Roman Empire. ' +
+      'Each chapter MUST HAVE at least 1000 words.',
+  })
+
+  console.log(text)
+  console.log()
+  console.log('Usage:', usage)
+  console.log('# of steps:', steps.length)
+}
+
+buildProgram('qwen2.5', main).catch(console.error)

examples/ai-core/src/generate-text/ollama-multi-step.ts

@@ -7,7 +7,7 @@ import { z } from 'zod'
 import { buildProgram } from '../tools/command'
 
 async function main(model: Parameters<typeof ollama>[0]) {
-  const { text, usage } = await generateText({
+  await generateText({
     maxSteps: 5,
     model: ollama(model),
     onStepFinish: (step) => {

examples/ai-core/src/registry/setup-registry.ts

@@ -7,8 +7,8 @@ import { ollama } from 'ollama-ai-provider'
 const myOllama = customProvider({
   fallbackProvider: ollama,
   languageModels: {
-    multimodal: ollama('llava'),
-    text: ollama('llama3.1'),
+    multimodal: ollama('llama3.2-vision'),
+    text: ollama('llama3.2'),
   },
 })
 

examples/ai-core/src/stream-text/ollama-chatbot.ts

@@ -2,8 +2,9 @@
 
 import * as readline from 'node:readline/promises'
 
-import { CoreMessage, streamText } from 'ai'
+import { CoreMessage, streamText, tool } from 'ai'
 import { ollama } from 'ollama-ai-provider'
+import { z } from 'zod'
 
 import { buildProgram } from '../tools/command'
 
@@ -21,20 +22,33 @@ async function main(model: Parameters<typeof ollama>[0]) {
     messages.push({ content: userInput, role: 'user' })
 
     const result = await streamText({
+      maxSteps: 5,
       messages,
       model: ollama(model),
-      system: `You are a helpful, respectful and honest assistant.`,
+      tools: {
+        weather: tool({
+          description: 'Get the weather in a location',
+          execute: async ({ location }) => ({
+            location,
+            temperature: 72 + Math.floor(Math.random() * 21) - 10,
+          }),
+          parameters: z.object({
+            location: z
+              .string()
+              .describe('The location to get the weather for'),
+          }),
+        }),
+      },
     })
 
-    let fullResponse = ''
     process.stdout.write('\nAssistant: ')
     for await (const delta of result.textStream) {
-      fullResponse += delta
       process.stdout.write(delta)
     }
     process.stdout.write('\n\n')
 
-    messages.push({ content: fullResponse, role: 'assistant' })
+    // eslint-disable-next-line unicorn/no-await-expression-member
+    messages.push(...(await result.response).messages)
   }
 }
 

examples/ai-core/src/tools/command.ts

@@ -3,8 +3,8 @@ import { ollama } from 'ollama-ai-provider'
 
 export async function buildProgram(
   defaultModel:
-    | Parameters<typeof ollama>[0]
-    | Parameters<typeof ollama.embedding>[0],
+    | Parameters<typeof ollama.languageModel>[0]
+    | Parameters<typeof ollama.textEmbeddingModel>[0],
   action: (model: string) => Promise<void>,
 ) {
   const program = new Command()

package.json

@@ -37,6 +37,9 @@
     "vite-tsconfig-paths": "^4.3.2",
     "vitest": "^2.1.3"
   },
+  "engines": {
+    "node": "^18.0.0 || ^20.0.0"
+  },
   "homepage": "https://github.com/sgomez/ollama-ai-provider",
   "repository": {
     "type": "git",
@@ -48,5 +51,5 @@
   "keywords": [
     "ai"
   ],
-  "packageManager": "pnpm@9.6.0"
+  "packageManager": "pnpm@9.12.3"
 }

packages/ollama/package.json

@@ -26,15 +26,15 @@
   "author": "Sergio Gómez Bachiller <[email protected]>",
   "license": "Apache-2.0",
   "dependencies": {
-    "@ai-sdk/provider": "0.0.26",
-    "@ai-sdk/provider-utils": "1.0.22",
+    "@ai-sdk/provider": "^1.0.0",
+    "@ai-sdk/provider-utils": "^2.0.0",
     "partial-json": "0.1.7"
   },
   "devDependencies": {
     "@edge-runtime/vm": "^3.2.0",
     "@types/node": "^18.19.56",
     "tsup": "^8.3.0",
-    "typescript": "5.5.4",
+    "typescript": "5.6.3",
     "zod": "3.23.8"
   },
   "peerDependencies": {

packages/ollama/src/index.ts

@@ -1,3 +1,2 @@
-export * from './ollama-facade'
 export type { OllamaProvider, OllamaProviderSettings } from './ollama-provider'
 export { createOllama, ollama } from './ollama-provider'

packages/ollama/src/ollama-chat-language-model.ts

@@ -98,10 +98,10 @@ export class OllamaChatLanguageModel implements LanguageModelV1 {
         repeat_last_n: this.settings.repeatLastN,
         repeat_penalty: this.settings.repeatPenalty,
         seed,
-        stop: this.settings.stop ?? stopSequences,
+        stop: stopSequences,
         temperature,
         tfs_z: this.settings.tfsZ,
-        top_k: this.settings.topK ?? topK,
+        top_k: topK,
         top_p: topP,
         typical_p: this.settings.typicalP,
         use_mlock: this.settings.useMlock,

packages/ollama/src/ollama-chat-settings.ts

@@ -1,4 +1,6 @@
 export type OllamaChatModelId =
+  | 'athene-v2'
+  | 'athene-v2:72b'
   | 'aya-expanse'
   | 'aya-expanse:8b'
   | 'aya-expanse:32b'
@@ -92,6 +94,9 @@ export type OllamaChatModelId =
   | 'nemotron:70b'
   | 'nemotron-mini'
   | 'nemotron-mini:4b'
+  | 'opencoder'
+  | 'opencoder:1.5b'
+  | 'opencoder:8b'
   | 'phi3'
   | 'phi3:3.8b'
   | 'phi3:14b'
@@ -117,8 +122,12 @@ export type OllamaChatModelId =
   | 'qwen2.5:32b'
   | 'qwen2.5:72b'
   | 'qwen2.5-coder'
+  | 'qwen2.5-coder:0.5b'
   | 'qwen2.5-coder:1.5b'
+  | 'qwen2.5-coder:3b'
   | 'qwen2.5-coder:7b'
+  | 'qwen2.5-coder:14b'
+  | 'qwen2.5-coder:32b'
   | 'shieldgemma'
   | 'shieldgemma:2b'
   | 'shieldgemma:9b'
@@ -127,6 +136,9 @@ export type OllamaChatModelId =
   | 'smollm:135m'
   | 'smollm:360m'
   | 'smollm:1.7b'
+  | 'tulu3'
+  | 'tulu3:8b'
+  | 'tulu3:70b'
   | (string & NonNullable<unknown>)
 
 export interface OllamaChatSettings {
@@ -224,28 +236,12 @@ export interface OllamaChatSettings {
    */
   repeatPenalty?: number
 
-  /**
-   * Sets the stop sequences to use. When this pattern is encountered the LLM will stop generating text and return.
-   * Multiple stop patterns may be set by specifying multiple separate `stop` parameters in a modelfile.
-   *
-   * @deprecated Use `stopSequences` from AI SDK functions.
-   */
-  stop?: string
-
   /**
    * Tail free sampling is used to reduce the impact of less probable tokens from the output. A higher value (e.g., 2.0)
    * will reduce the impact more, while a value of 1.0 disables this setting. (default: 1)
    */
   tfsZ?: number
 
-  /**
-   * Reduces the probability of generating nonsense. A higher value (e.g. 100) will give more diverse answers, while a
-   * lower value (e.g. 10) will be more conservative. (Default: 40)
-   *
-   * @deprecated Use `topK` from AI SDK functions.
-   */
-  topK?: number
-
   /**
    * Controls the "typical" sampling probability. (Default: 1.0)
    */