ai-sdk-ollama 4.2.0 → 4.4.0

package/CHANGELOG.md

@@ -1,5 +1,28 @@
 # Changelog
 
+## 4.4.0
+
+### Minor Changes
+
+- 004f6b4: Add `ollama.evaluationModel()` for Ollama decision models such as `nimble` and `tev1`. Use it with the AI SDK's `experimental_evaluate` to get typed choice, boolean and score answers with probabilities from one local call.
+
+  - `OllamaClient` accepts an optional `systemone()` method.
+  - `generateText` and `streamText` restate the latest user message, text parts included, when they synthesise a reply from tool results.
+  - JSON repair serialises object values for string fields as JSON.
+  - Supports `ai` 7.0.103+ and `ollama` 0.6.4.
+
+## 4.3.0
+
+### Minor Changes
+
+- 371f515: Track AI SDK 7.0.95 and open the provider up to custom Ollama clients.
+
+  - The `ai` peer range moves to `^7.0.95`, with `@ai-sdk/provider` and `@ai-sdk/provider-utils` updated to match.
+  - `OllamaClient` is a structural contract, so the official client or a custom adapter can be injected in both Node.js and browser builds. Streaming methods use the newly exported `AbortableStream` type.
+  - Per-request `AbortSignal`s are forwarded to the client, and an aborted request settles as a rejection across the reliability retry and fallback paths.
+  - Returned `message.thinking` is surfaced as AI SDK reasoning independently of the request's `think` setting, including on terminal stream chunks, and streamed text and reasoning blocks close symmetrically.
+  - Output usage carries the total token count Ollama reports and leaves the text/reasoning split unset, matching what the API provides.
+
 ## 4.2.0
 
 ### Minor Changes

package/README.md

@@ -154,6 +154,7 @@ export OLLAMA_API_KEY="your_api_key_here"
     - [Combining Tools with Structured Output](#combining-tools-with-structured-output)
     - [Simple and Predictable](#simple-and-predictable)
   - [Reranking](#reranking)
+  - [Decision Models](#decision-models)
   - [Streaming Utilities](#streaming-utilities)
     - [Smooth Stream](#smooth-stream)
     - [Partial JSON Parsing](#partial-json-parsing)
@@ -481,6 +482,46 @@ ranking.forEach((item, i) => {
 
 **Recommended Models**: `embeddinggemma` (best score separation), `nomic-embed-text`, `bge-m3`
 
+## Decision Models
+
+Ollama 0.35 serves decision models (`nimble`, `tev1`, `tev1:0.8b`) on `/v1/systemone`. You ask typed questions about some state and get each answer with its probabilities in one local call. Use them for triage, routing and moderation through the AI SDK's `experimental_evaluate`:
+
+```typescript
+import { experimental_evaluate as evaluate } from 'ai';
+import { ollama } from 'ai-sdk-ollama';
+
+const { answers } = await evaluate({
+  model: ollama.evaluationModel('nimble'),
+  state: { ticket: 'I was charged twice. Please refund the extra payment.' },
+  questions: {
+    team: {
+      type: 'choice',
+      instructions: 'Which team should handle this ticket?',
+      criteria: {
+        billing: 'Payments and refunds',
+        technical: null,
+        other: null,
+      },
+    },
+    refund: {
+      type: 'boolean',
+      instructions: 'Does the customer ask for a refund?',
+    },
+    urgency: {
+      type: 'score',
+      instructions: 'How urgent is this ticket?',
+      criteria: ['Routine', 'Soon', 'Urgent'],
+    },
+  },
+});
+
+answers.team.choice; // 'billing'
+answers.refund.probability; // P(true), e.g. 0.99
+answers.urgency.score; // 0 to 2, e.g. 0.8
+```
+
+Ollama names boolean questions `noul`, and the provider maps them for you. Ollama's per-question `confidence` lands in `providerMetadata.ollama.confidence`. Pull a model first with `ollama pull nimble`.
+
 ## Streaming Utilities
 
 ### Smooth Stream
@@ -740,7 +781,9 @@ const ollama = createOllama({ apiKey: process.env.OLLAMA_API_KEY });
 
 ### Using Existing Ollama Client
 
-You can also pass an existing Ollama client instance to reuse your configuration:
+You can also pass an existing Ollama client instance to reuse your configuration.
+The provider accepts the exported structural `OllamaClient` interface, so you
+can inject the official client or a custom adapter in Node.js and browser builds.
 
 ```typescript
 import { Ollama } from 'ollama';
@@ -756,13 +799,36 @@ const existingClient = new Ollama({
 const ollamaSdk = createOllama({ client: existingClient });
 
 // Use both clients as needed
-await ollamaRaw.list(); // Direct Ollama operations
+await existingClient.list(); // Direct Ollama operations
 const { text } = await generateText({
   model: ollamaSdk('llama3.2'),
   prompt: 'Hello!',
 });
 ```
 
+#### Custom client adapters
+
+`OllamaClient` is structural, so any object with the right shape works — no
+`Ollama` instance required. Streaming methods return an `AbortableStream`, an
+async iterable with an `abort()`:
+
+```typescript
+import type { AbortableStream, OllamaClient } from 'ai-sdk-ollama';
+import { createOllama } from 'ai-sdk-ollama';
+
+const client: OllamaClient = {
+  chat: myChat, // returns a ChatResponse, or an AbortableStream of them
+  embed: myEmbed,
+  webSearch: myWebSearch,
+  webFetch: myWebFetch,
+};
+
+const ollama = createOllama({ client });
+```
+
+Per-request cancellation is forwarded to the client as an `AbortSignal`, and an
+aborted request rejects rather than resolving with a retried or fallback result.
+
 ### Structured Output
 
 ```typescript
@@ -918,39 +984,38 @@ const { output: custom } = await generateText({
 
 ### Reasoning Support
 
-Some models like DeepSeek-R1 support reasoning (chain-of-thought) output. Enable this feature to see the model's thinking process:
+Use `think` to request reasoning generation from models that support it. Returned `message.thinking` is exposed as AI SDK reasoning, separately from final text, in both generation results and streams (including terminal chunks).
 
 ```typescript
 import { ollama } from 'ai-sdk-ollama';
 import { generateText } from 'ai';
 
 // Enable reasoning for models that support it (e.g., deepseek-r1:7b)
-const model = ollama('deepseek-r1:7b', { reasoning: true });
+const model = ollama('deepseek-r1:7b', { think: true });
 
 // Generate text with reasoning
-const { text } = await generateText({
+const { text, reasoningText } = await generateText({
   model,
   prompt:
     'Solve: If I have 3 boxes, each with 4 smaller boxes, and each smaller box has 5 items, how many items total?',
 });
 
 console.log('Answer:', text);
-// DeepSeek-R1 includes reasoning in the output with <think> tags:
-// <think>
-// First, I'll calculate the number of smaller boxes: 3 × 4 = 12
-// Then, the total items: 12 × 5 = 60
-// </think>
-// You have 60 items in total.
+console.log('Reasoning:', reasoningText);
 
 // Compare with reasoning disabled
-const modelNoReasoning = ollama('deepseek-r1:7b', { reasoning: false });
+const modelNoReasoning = ollama('deepseek-r1:7b', { think: false });
 const { text: noReasoningText } = await generateText({
   model: modelNoReasoning,
   prompt: 'Calculate 3 × 4 × 5',
 });
-// Output: 60 (without showing the thinking process)
+console.log('Answer:', noReasoningText);
 ```
 
+`think` controls the request, not filtering of the response. If the server returns thinking with `think` omitted or `false`, the adapter still preserves it as reasoning; it does not turn it into final text. If no thinking is returned, no reasoning content is added.
+
+Ollama supplies a combined output token count. The adapter preserves that total and the raw counters, but leaves the text/reasoning token breakdown unknown instead of attributing all output tokens to text.
+
 **Recommended Reasoning Models**:
 
 - `deepseek-r1:7b` - Balanced performance and reasoning capability (5GB)

package/dist/functions/generate-text.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"generate-text.d.ts","sourceRoot":"","sources":["../../src/functions/generate-text.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAEH,OAAO,EAAE,YAAY,IAAI,aAAa,EAAe,MAAM,IAAI,CAAC;AAqBhE;;GAEG;AACH,MAAM,WAAW,eAAe;IAC9B;;;OAGG;IACH,eAAe,CAAC,EAAE,OAAO,CAAC;IAE1B;;OAEG;IACH,eAAe,CAAC,EAAE,MAAM,CAAC;IAEzB;;;OAGG;IACH,oBAAoB,CAAC,EAAE,MAAM,CAAC;IAE9B;;;OAGG;IACH,iBAAiB,CAAC,EAAE,MAAM,CAAC;IAE3B;;;;;;;;;;;OAWG;IACH,+BAA+B,CAAC,EAAE,OAAO,CAAC;CAC3C;AAED;;GAEG;AACH,MAAM,MAAM,mBAAmB,GAAG,UAAU,CAAC,OAAO,aAAa,CAAC,CAAC,CAAC,CAAC,GAAG;IACtE;;OAEG;IACH,eAAe,CAAC,EAAE,eAAe,CAAC;CACnC,CAAC;AAEF;;;;;;;GAOG;AACH,wBAAsB,YAAY,CAChC,OAAO,EAAE,mBAAmB,GAC3B,OAAO,CAAC,OAAO,CAAC,UAAU,CAAC,OAAO,aAAa,CAAC,CAAC,CAAC,CA4NpD"}
+{"version":3,"file":"generate-text.d.ts","sourceRoot":"","sources":["../../src/functions/generate-text.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAEH,OAAO,EAAE,YAAY,IAAI,aAAa,EAAe,MAAM,IAAI,CAAC;AAsBhE;;GAEG;AACH,MAAM,WAAW,eAAe;IAC9B;;;OAGG;IACH,eAAe,CAAC,EAAE,OAAO,CAAC;IAE1B;;OAEG;IACH,eAAe,CAAC,EAAE,MAAM,CAAC;IAEzB;;;OAGG;IACH,oBAAoB,CAAC,EAAE,MAAM,CAAC;IAE9B;;;OAGG;IACH,iBAAiB,CAAC,EAAE,MAAM,CAAC;IAE3B;;;;;;;;;;;OAWG;IACH,+BAA+B,CAAC,EAAE,OAAO,CAAC;CAC3C;AAED;;GAEG;AACH,MAAM,MAAM,mBAAmB,GAAG,UAAU,CAAC,OAAO,aAAa,CAAC,CAAC,CAAC,CAAC,GAAG;IACtE;;OAEG;IACH,eAAe,CAAC,EAAE,eAAe,CAAC;CACnC,CAAC;AAEF;;;;;;;GAOG;AACH,wBAAsB,YAAY,CAChC,OAAO,EAAE,mBAAmB,GAC3B,OAAO,CAAC,OAAO,CAAC,UAAU,CAAC,OAAO,aAAa,CAAC,CAAC,CAAC,CAyNpD"}

package/dist/functions/stream-text.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"stream-text.d.ts","sourceRoot":"","sources":["../../src/functions/stream-text.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,UAAU,IAAI,WAAW,EAAe,MAAM,IAAI,CAAC;AAE5D,KAAK,mBAAmB,GAAG,UAAU,CAAC,OAAO,WAAW,CAAC,CAAC,CAAC,CAAC,CAAC;AAsC7D,MAAM,MAAM,iBAAiB,GAAG,mBAAmB,GAAG;IACpD,eAAe,CAAC,EAAE;QAChB;;WAEG;QACH,iBAAiB,CAAC,EAAE,OAAO,CAAC;QAC5B;;WAEG;QACH,wBAAwB,CAAC,EAAE,OAAO,CAAC;QACnC;;WAEG;QACH,eAAe,CAAC,EAAE,MAAM,CAAC;QACzB;;WAEG;QACH,gBAAgB,CAAC,EAAE,MAAM,CAAC;KAC3B,CAAC;CACH,CAAC;AAUF,wBAAsB,UAAU,CAC9B,OAAO,EAAE,iBAAiB,GACzB,OAAO,CAAC,OAAO,CAAC,UAAU,CAAC,OAAO,WAAW,CAAC,CAAC,CAAC,CAsclD"}
+{"version":3,"file":"stream-text.d.ts","sourceRoot":"","sources":["../../src/functions/stream-text.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,UAAU,IAAI,WAAW,EAAe,MAAM,IAAI,CAAC;AAG5D,KAAK,mBAAmB,GAAG,UAAU,CAAC,OAAO,WAAW,CAAC,CAAC,CAAC,CAAC,CAAC;AAsC7D,MAAM,MAAM,iBAAiB,GAAG,mBAAmB,GAAG;IACpD,eAAe,CAAC,EAAE;QAChB;;WAEG;QACH,iBAAiB,CAAC,EAAE,OAAO,CAAC;QAC5B;;WAEG;QACH,wBAAwB,CAAC,EAAE,OAAO,CAAC;QACnC;;WAEG;QACH,eAAe,CAAC,EAAE,MAAM,CAAC;QACzB;;WAEG;QACH,gBAAgB,CAAC,EAAE,MAAM,CAAC;KAC3B,CAAC;CACH,CAAC;AAUF,wBAAsB,UAAU,CAC9B,OAAO,EAAE,iBAAiB,GACzB,OAAO,CAAC,OAAO,CAAC,UAAU,CAAC,OAAO,WAAW,CAAC,CAAC,CAAC,CAkclD"}