modelfusion 0.105.0 → 0.107.0

package/model-provider/cohere/CohereTextEmbeddingModel.d.ts

@@ -74,8 +74,8 @@ export declare class CohereTextEmbeddingModel extends AbstractModel<CohereTextEm
     get settingsForEvent(): Partial<CohereTextEmbeddingModelSettings>;
     doEmbedValues(texts: string[], options?: FunctionOptions): Promise<{
         response: {
-            texts: string[];
             embeddings: number[][];
+            texts: string[];
             id: string;
             meta: {
                 api_version: {
@@ -109,8 +109,8 @@ declare const cohereTextEmbeddingResponseSchema: z.ZodObject<{
         };
     }>;
 }, "strip", z.ZodTypeAny, {
-    texts: string[];
     embeddings: number[][];
+    texts: string[];
     id: string;
     meta: {
         api_version: {
@@ -118,8 +118,8 @@ declare const cohereTextEmbeddingResponseSchema: z.ZodObject<{
         };
     };
 }, {
-    texts: string[];
     embeddings: number[][];
+    texts: string[];
     id: string;
     meta: {
         api_version: {

package/model-provider/cohere/CohereTextGenerationModel.d.ts

@@ -66,8 +66,8 @@ export declare class CohereTextGenerationModel extends AbstractModel<CohereTextG
     get settingsForEvent(): Partial<CohereTextGenerationModelSettings>;
     doGenerateTexts(prompt: string, options?: FunctionOptions): Promise<{
         response: {
-            prompt: string;
             id: string;
+            prompt: string;
             generations: {
                 text: string;
                 id: string;
@@ -90,8 +90,8 @@ export declare class CohereTextGenerationModel extends AbstractModel<CohereTextG
         is_finished: false;
     } | {
         response: {
-            prompt: string;
             id: string;
+            prompt: string;
             generations: {
                 text: string;
                 id: string;
@@ -155,8 +155,8 @@ declare const cohereTextGenerationResponseSchema: z.ZodObject<{
         };
     }>>;
 }, "strip", z.ZodTypeAny, {
-    prompt: string;
     id: string;
+    prompt: string;
     generations: {
         text: string;
         id: string;
@@ -168,8 +168,8 @@ declare const cohereTextGenerationResponseSchema: z.ZodObject<{
         };
     } | undefined;
 }, {
-    prompt: string;
     id: string;
+    prompt: string;
     generations: {
         text: string;
         id: string;
@@ -193,8 +193,8 @@ export declare const CohereTextGenerationResponseFormat: {
     json: {
         stream: boolean;
         handler: ResponseHandler<{
-            prompt: string;
             id: string;
+            prompt: string;
             generations: {
                 text: string;
                 id: string;
@@ -220,8 +220,8 @@ export declare const CohereTextGenerationResponseFormat: {
             is_finished: false;
         } | {
             response: {
-                prompt: string;
                 id: string;
+                prompt: string;
                 generations: {
                     text: string;
                     id: string;

package/model-provider/llamacpp/LlamaCppBakLLaVA1PromptTemplate.d.ts

@@ -1,11 +1,11 @@
 import { TextGenerationPromptTemplate } from "../../model-function/generate-text/TextGenerationPromptTemplate.js";
 import { ChatPrompt } from "../../model-function/generate-text/prompt-template/ChatPrompt.js";
 import { InstructionPrompt } from "../../model-function/generate-text/prompt-template/InstructionPrompt.js";
-import { LlamaCppTextGenerationPrompt } from "./LlamaCppTextGenerationModel.js";
+import { LlamaCppCompletionPrompt } from "./LlamaCppCompletionModel.js";
 /**
  * BakLLaVA 1 uses a Vicuna 1 prompt. This mapping combines it with the LlamaCpp prompt structure.
  *
  * @see https://github.com/SkunkworksAI/BakLLaVA
  */
-export declare function instruction(): TextGenerationPromptTemplate<InstructionPrompt, LlamaCppTextGenerationPrompt>;
-export declare function chat(): TextGenerationPromptTemplate<ChatPrompt, LlamaCppTextGenerationPrompt>;
+export declare function instruction(): TextGenerationPromptTemplate<InstructionPrompt, LlamaCppCompletionPrompt>;
+export declare function chat(): TextGenerationPromptTemplate<ChatPrompt, LlamaCppCompletionPrompt>;

package/model-provider/llamacpp/{LlamaCppTextGenerationModel.cjs → LlamaCppCompletionModel.cjs}

@@ -1,6 +1,6 @@
 "use strict";
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.LlamaCppTextGenerationResponseFormat = exports.LlamaCppTextGenerationModel = void 0;
+exports.LlamaCppCompletionResponseFormat = exports.LlamaCppCompletionModel = void 0;
 const zod_1 = require("zod");
 const callWithRetryAndThrottle_js_1 = require("../../core/api/callWithRetryAndThrottle.cjs");
 const postToApi_js_1 = require("../../core/api/postToApi.cjs");
@@ -14,7 +14,7 @@ const parseEventSourceStream_js_1 = require("../../util/streaming/parseEventSour
 const LlamaCppApiConfiguration_js_1 = require("./LlamaCppApiConfiguration.cjs");
 const LlamaCppError_js_1 = require("./LlamaCppError.cjs");
 const LlamaCppTokenizer_js_1 = require("./LlamaCppTokenizer.cjs");
-class LlamaCppTextGenerationModel extends AbstractModel_js_1.AbstractModel {
+class LlamaCppCompletionModel extends AbstractModel_js_1.AbstractModel {
     constructor(settings = {}) {
         super({ settings });
         Object.defineProperty(this, "provider", {
@@ -111,7 +111,7 @@ class LlamaCppTextGenerationModel extends AbstractModel_js_1.AbstractModel {
     async doGenerateTexts(prompt, options) {
         const response = await this.callAPI(prompt, {
             ...options,
-            responseFormat: exports.LlamaCppTextGenerationResponseFormat.json,
+            responseFormat: exports.LlamaCppCompletionResponseFormat.json,
         });
         return {
             response,
@@ -135,7 +135,7 @@ class LlamaCppTextGenerationModel extends AbstractModel_js_1.AbstractModel {
     doStreamText(prompt, options) {
         return this.callAPI(prompt, {
             ...options,
-            responseFormat: exports.LlamaCppTextGenerationResponseFormat.deltaIterable,
+            responseFormat: exports.LlamaCppCompletionResponseFormat.deltaIterable,
         });
     }
     extractTextDelta(delta) {
@@ -178,10 +178,10 @@ class LlamaCppTextGenerationModel extends AbstractModel_js_1.AbstractModel {
         });
     }
     withSettings(additionalSettings) {
-        return new LlamaCppTextGenerationModel(Object.assign({}, this.settings, additionalSettings));
+        return new LlamaCppCompletionModel(Object.assign({}, this.settings, additionalSettings));
     }
 }
-exports.LlamaCppTextGenerationModel = LlamaCppTextGenerationModel;
+exports.LlamaCppCompletionModel = LlamaCppCompletionModel;
 const llamaCppTextGenerationResponseSchema = zod_1.z.object({
     content: zod_1.z.string(),
     stop: zod_1.z.literal(true),
@@ -204,7 +204,7 @@ const llamaCppTextGenerationResponseSchema = zod_1.z.object({
         seed: zod_1.z.number(),
         stop: zod_1.z.array(zod_1.z.string()),
         stream: zod_1.z.boolean(),
-        temp: zod_1.z.number(),
+        temperature: zod_1.z.number().optional(), // optional for backwards compatibility
         tfs_z: zod_1.z.number(),
         top_k: zod_1.z.number(),
         top_p: zod_1.z.number(),
@@ -267,7 +267,7 @@ async function createLlamaCppFullDeltaIterableQueue(stream) {
     });
     return queue;
 }
-exports.LlamaCppTextGenerationResponseFormat = {
+exports.LlamaCppCompletionResponseFormat = {
     /**
      * Returns the response as a JSON object.
      */

package/model-provider/llamacpp/{LlamaCppTextGenerationModel.d.ts → LlamaCppCompletionModel.d.ts}

@@ -8,7 +8,7 @@ import { PromptTemplateTextStreamingModel } from "../../model-function/generate-
 import { TextGenerationModelSettings, TextStreamingModel } from "../../model-function/generate-text/TextGenerationModel.js";
 import { TextGenerationPromptTemplate } from "../../model-function/generate-text/TextGenerationPromptTemplate.js";
 import { LlamaCppTokenizer } from "./LlamaCppTokenizer.js";
-export interface LlamaCppTextGenerationModelSettings<CONTEXT_WINDOW_SIZE extends number | undefined> extends TextGenerationModelSettings {
+export interface LlamaCppCompletionModelSettings<CONTEXT_WINDOW_SIZE extends number | undefined> extends TextGenerationModelSettings {
     api?: ApiConfiguration;
     /**
      * Specify the context window size of the model that you have loaded in your
@@ -35,7 +35,7 @@ export interface LlamaCppTextGenerationModelSettings<CONTEXT_WINDOW_SIZE extends
     ignoreEos?: boolean;
     logitBias?: Array<[number, number | false]>;
 }
-export interface LlamaCppTextGenerationPrompt {
+export interface LlamaCppCompletionPrompt {
     /**
      * Text prompt. Images can be included through references such as `[img-ID]`, e.g. `[img-1]`.
      */
@@ -45,27 +45,27 @@ export interface LlamaCppTextGenerationPrompt {
      */
     images?: Record<number, string>;
 }
-export declare class LlamaCppTextGenerationModel<CONTEXT_WINDOW_SIZE extends number | undefined> extends AbstractModel<LlamaCppTextGenerationModelSettings<CONTEXT_WINDOW_SIZE>> implements TextStreamingModel<LlamaCppTextGenerationPrompt, LlamaCppTextGenerationModelSettings<CONTEXT_WINDOW_SIZE>> {
-    constructor(settings?: LlamaCppTextGenerationModelSettings<CONTEXT_WINDOW_SIZE>);
+export declare class LlamaCppCompletionModel<CONTEXT_WINDOW_SIZE extends number | undefined> extends AbstractModel<LlamaCppCompletionModelSettings<CONTEXT_WINDOW_SIZE>> implements TextStreamingModel<LlamaCppCompletionPrompt, LlamaCppCompletionModelSettings<CONTEXT_WINDOW_SIZE>> {
+    constructor(settings?: LlamaCppCompletionModelSettings<CONTEXT_WINDOW_SIZE>);
     readonly provider = "llamacpp";
     get modelName(): null;
     get contextWindowSize(): CONTEXT_WINDOW_SIZE;
     readonly tokenizer: LlamaCppTokenizer;
-    callAPI<RESPONSE>(prompt: LlamaCppTextGenerationPrompt, options: {
-        responseFormat: LlamaCppTextGenerationResponseFormatType<RESPONSE>;
+    callAPI<RESPONSE>(prompt: LlamaCppCompletionPrompt, options: {
+        responseFormat: LlamaCppCompletionResponseFormatType<RESPONSE>;
     } & FunctionOptions): Promise<RESPONSE>;
-    get settingsForEvent(): Partial<LlamaCppTextGenerationModelSettings<CONTEXT_WINDOW_SIZE>>;
-    countPromptTokens(prompt: LlamaCppTextGenerationPrompt): Promise<number>;
-    doGenerateTexts(prompt: LlamaCppTextGenerationPrompt, options?: FunctionOptions): Promise<{
+    get settingsForEvent(): Partial<LlamaCppCompletionModelSettings<CONTEXT_WINDOW_SIZE>>;
+    countPromptTokens(prompt: LlamaCppCompletionPrompt): Promise<number>;
+    doGenerateTexts(prompt: LlamaCppCompletionPrompt, options?: FunctionOptions): Promise<{
         response: {
-            stop: true;
             model: string;
-            prompt: string;
+            stop: true;
             content: string;
+            prompt: string;
             generation_settings: {
-                stop: string[];
                 model: string;
                 stream: boolean;
+                stop: string[];
                 seed: number;
                 mirostat: number;
                 frequency_penalty: number;
@@ -81,11 +81,11 @@ export declare class LlamaCppTextGenerationModel<CONTEXT_WINDOW_SIZE extends num
                 presence_penalty: number;
                 repeat_last_n: number;
                 repeat_penalty: number;
-                temp: number;
                 tfs_z: number;
                 top_k: number;
                 top_p: number;
                 typical_p: number;
+                temperature?: number | undefined;
             };
             stopped_eos: boolean;
             stopped_limit: boolean;
@@ -116,15 +116,15 @@ export declare class LlamaCppTextGenerationModel<CONTEXT_WINDOW_SIZE extends num
             totalTokens: number;
         };
     }>;
-    doStreamText(prompt: LlamaCppTextGenerationPrompt, options?: FunctionOptions): Promise<AsyncIterable<Delta<{
-        stop: true;
+    doStreamText(prompt: LlamaCppCompletionPrompt, options?: FunctionOptions): Promise<AsyncIterable<Delta<{
         model: string;
-        prompt: string;
+        stop: true;
         content: string;
+        prompt: string;
         generation_settings: {
-            stop: string[];
             model: string;
             stream: boolean;
+            stop: string[];
             seed: number;
             mirostat: number;
             frequency_penalty: number;
@@ -140,11 +140,11 @@ export declare class LlamaCppTextGenerationModel<CONTEXT_WINDOW_SIZE extends num
             presence_penalty: number;
             repeat_last_n: number;
             repeat_penalty: number;
-            temp: number;
             tfs_z: number;
             top_k: number;
             top_p: number;
             typical_p: number;
+            temperature?: number | undefined;
         };
         stopped_eos: boolean;
         stopped_limit: boolean;
@@ -169,16 +169,16 @@ export declare class LlamaCppTextGenerationModel<CONTEXT_WINDOW_SIZE extends num
         content: string;
     }>>>;
     extractTextDelta(delta: unknown): string;
-    withTextPrompt(): PromptTemplateTextStreamingModel<string, LlamaCppTextGenerationPrompt, LlamaCppTextGenerationModelSettings<CONTEXT_WINDOW_SIZE>, this>;
+    withTextPrompt(): PromptTemplateTextStreamingModel<string, LlamaCppCompletionPrompt, LlamaCppCompletionModelSettings<CONTEXT_WINDOW_SIZE>, this>;
     /**
      * Maps the prompt for a text version of the Llama.cpp prompt template (without image support).
      */
-    withTextPromptTemplate<INPUT_PROMPT>(promptTemplate: TextGenerationPromptTemplate<INPUT_PROMPT, string>): PromptTemplateTextStreamingModel<INPUT_PROMPT, string, LlamaCppTextGenerationModelSettings<CONTEXT_WINDOW_SIZE>, PromptTemplateTextStreamingModel<string, LlamaCppTextGenerationPrompt, LlamaCppTextGenerationModelSettings<CONTEXT_WINDOW_SIZE>, this>>;
+    withTextPromptTemplate<INPUT_PROMPT>(promptTemplate: TextGenerationPromptTemplate<INPUT_PROMPT, string>): PromptTemplateTextStreamingModel<INPUT_PROMPT, string, LlamaCppCompletionModelSettings<CONTEXT_WINDOW_SIZE>, PromptTemplateTextStreamingModel<string, LlamaCppCompletionPrompt, LlamaCppCompletionModelSettings<CONTEXT_WINDOW_SIZE>, this>>;
     /**
      * Maps the prompt for the full Llama.cpp prompt template (incl. image support).
      */
-    withPromptTemplate<INPUT_PROMPT>(promptTemplate: TextGenerationPromptTemplate<INPUT_PROMPT, LlamaCppTextGenerationPrompt>): PromptTemplateTextStreamingModel<INPUT_PROMPT, LlamaCppTextGenerationPrompt, LlamaCppTextGenerationModelSettings<CONTEXT_WINDOW_SIZE>, this>;
-    withSettings(additionalSettings: Partial<LlamaCppTextGenerationModelSettings<CONTEXT_WINDOW_SIZE>>): this;
+    withPromptTemplate<INPUT_PROMPT>(promptTemplate: TextGenerationPromptTemplate<INPUT_PROMPT, LlamaCppCompletionPrompt>): PromptTemplateTextStreamingModel<INPUT_PROMPT, LlamaCppCompletionPrompt, LlamaCppCompletionModelSettings<CONTEXT_WINDOW_SIZE>, this>;
+    withSettings(additionalSettings: Partial<LlamaCppCompletionModelSettings<CONTEXT_WINDOW_SIZE>>): this;
 }
 declare const llamaCppTextGenerationResponseSchema: z.ZodObject<{
     content: z.ZodString;
@@ -202,15 +202,15 @@ declare const llamaCppTextGenerationResponseSchema: z.ZodObject<{
         seed: z.ZodNumber;
         stop: z.ZodArray<z.ZodString, "many">;
         stream: z.ZodBoolean;
-        temp: z.ZodNumber;
+        temperature: z.ZodOptional<z.ZodNumber>;
         tfs_z: z.ZodNumber;
         top_k: z.ZodNumber;
         top_p: z.ZodNumber;
         typical_p: z.ZodNumber;
     }, "strip", z.ZodTypeAny, {
-        stop: string[];
         model: string;
         stream: boolean;
+        stop: string[];
         seed: number;
         mirostat: number;
         frequency_penalty: number;
@@ -226,15 +226,15 @@ declare const llamaCppTextGenerationResponseSchema: z.ZodObject<{
         presence_penalty: number;
         repeat_last_n: number;
         repeat_penalty: number;
-        temp: number;
         tfs_z: number;
         top_k: number;
         top_p: number;
         typical_p: number;
+        temperature?: number | undefined;
     }, {
-        stop: string[];
         model: string;
         stream: boolean;
+        stop: string[];
         seed: number;
         mirostat: number;
         frequency_penalty: number;
@@ -250,11 +250,11 @@ declare const llamaCppTextGenerationResponseSchema: z.ZodObject<{
         presence_penalty: number;
         repeat_last_n: number;
         repeat_penalty: number;
-        temp: number;
         tfs_z: number;
         top_k: number;
         top_p: number;
         typical_p: number;
+        temperature?: number | undefined;
     }>;
     model: z.ZodString;
     prompt: z.ZodString;
@@ -295,14 +295,14 @@ declare const llamaCppTextGenerationResponseSchema: z.ZodObject<{
     tokens_predicted: z.ZodNumber;
     truncated: z.ZodBoolean;
 }, "strip", z.ZodTypeAny, {
-    stop: true;
     model: string;
-    prompt: string;
+    stop: true;
     content: string;
+    prompt: string;
     generation_settings: {
-        stop: string[];
         model: string;
         stream: boolean;
+        stop: string[];
         seed: number;
         mirostat: number;
         frequency_penalty: number;
@@ -318,11 +318,11 @@ declare const llamaCppTextGenerationResponseSchema: z.ZodObject<{
         presence_penalty: number;
         repeat_last_n: number;
         repeat_penalty: number;
-        temp: number;
         tfs_z: number;
         top_k: number;
         top_p: number;
         typical_p: number;
+        temperature?: number | undefined;
     };
     stopped_eos: boolean;
     stopped_limit: boolean;
@@ -343,14 +343,14 @@ declare const llamaCppTextGenerationResponseSchema: z.ZodObject<{
     tokens_predicted: number;
     truncated: boolean;
 }, {
-    stop: true;
     model: string;
-    prompt: string;
+    stop: true;
     content: string;
+    prompt: string;
     generation_settings: {
-        stop: string[];
         model: string;
         stream: boolean;
+        stop: string[];
         seed: number;
         mirostat: number;
         frequency_penalty: number;
@@ -366,11 +366,11 @@ declare const llamaCppTextGenerationResponseSchema: z.ZodObject<{
         presence_penalty: number;
         repeat_last_n: number;
         repeat_penalty: number;
-        temp: number;
         tfs_z: number;
         top_k: number;
         top_p: number;
         typical_p: number;
+        temperature?: number | undefined;
     };
     stopped_eos: boolean;
     stopped_limit: boolean;
@@ -393,14 +393,14 @@ declare const llamaCppTextGenerationResponseSchema: z.ZodObject<{
 }>;
 export type LlamaCppTextGenerationResponse = z.infer<typeof llamaCppTextGenerationResponseSchema>;
 declare const llamaCppTextStreamChunkSchema: import("../../core/schema/ZodSchema.js").ZodSchema<{
-    stop: true;
     model: string;
-    prompt: string;
+    stop: true;
     content: string;
+    prompt: string;
     generation_settings: {
-        stop: string[];
         model: string;
         stream: boolean;
+        stop: string[];
         seed: number;
         mirostat: number;
         frequency_penalty: number;
@@ -416,11 +416,11 @@ declare const llamaCppTextStreamChunkSchema: import("../../core/schema/ZodSchema
         presence_penalty: number;
         repeat_last_n: number;
         repeat_penalty: number;
-        temp: number;
         tfs_z: number;
         top_k: number;
         top_p: number;
         typical_p: number;
+        temperature?: number | undefined;
     };
     stopped_eos: boolean;
     stopped_limit: boolean;
@@ -445,25 +445,25 @@ declare const llamaCppTextStreamChunkSchema: import("../../core/schema/ZodSchema
     content: string;
 }>;
 export type LlamaCppTextStreamChunk = (typeof llamaCppTextStreamChunkSchema)["_type"];
-export type LlamaCppTextGenerationResponseFormatType<T> = {
+export type LlamaCppCompletionResponseFormatType<T> = {
     stream: boolean;
     handler: ResponseHandler<T>;
 };
-export declare const LlamaCppTextGenerationResponseFormat: {
+export declare const LlamaCppCompletionResponseFormat: {
     /**
      * Returns the response as a JSON object.
      */
     json: {
         stream: false;
         handler: ResponseHandler<{
-            stop: true;
             model: string;
-            prompt: string;
+            stop: true;
             content: string;
+            prompt: string;
             generation_settings: {
-                stop: string[];
                 model: string;
                 stream: boolean;
+                stop: string[];
                 seed: number;
                 mirostat: number;
                 frequency_penalty: number;
@@ -479,11 +479,11 @@ export declare const LlamaCppTextGenerationResponseFormat: {
                 presence_penalty: number;
                 repeat_last_n: number;
                 repeat_penalty: number;
-                temp: number;
                 tfs_z: number;
                 top_k: number;
                 top_p: number;
                 typical_p: number;
+                temperature?: number | undefined;
             };
             stopped_eos: boolean;
             stopped_limit: boolean;
@@ -514,14 +514,14 @@ export declare const LlamaCppTextGenerationResponseFormat: {
         handler: ({ response }: {
             response: Response;
         }) => Promise<AsyncIterable<Delta<{
-            stop: true;
             model: string;
-            prompt: string;
+            stop: true;
             content: string;
+            prompt: string;
             generation_settings: {
-                stop: string[];
                 model: string;
                 stream: boolean;
+                stop: string[];
                 seed: number;
                 mirostat: number;
                 frequency_penalty: number;
@@ -537,11 +537,11 @@ export declare const LlamaCppTextGenerationResponseFormat: {
                 presence_penalty: number;
                 repeat_last_n: number;
                 repeat_penalty: number;
-                temp: number;
                 tfs_z: number;
                 top_k: number;
                 top_p: number;
                 typical_p: number;
+                temperature?: number | undefined;
             };
             stopped_eos: boolean;
             stopped_limit: boolean;

package/model-provider/llamacpp/{LlamaCppTextGenerationModel.js → LlamaCppCompletionModel.js}

@@ -11,7 +11,7 @@ import { parseEventSourceStream } from "../../util/streaming/parseEventSourceStr
 import { LlamaCppApiConfiguration } from "./LlamaCppApiConfiguration.js";
 import { failedLlamaCppCallResponseHandler } from "./LlamaCppError.js";
 import { LlamaCppTokenizer } from "./LlamaCppTokenizer.js";
-export class LlamaCppTextGenerationModel extends AbstractModel {
+export class LlamaCppCompletionModel extends AbstractModel {
     constructor(settings = {}) {
         super({ settings });
         Object.defineProperty(this, "provider", {
@@ -108,7 +108,7 @@ export class LlamaCppTextGenerationModel extends AbstractModel {
     async doGenerateTexts(prompt, options) {
         const response = await this.callAPI(prompt, {
             ...options,
-            responseFormat: LlamaCppTextGenerationResponseFormat.json,
+            responseFormat: LlamaCppCompletionResponseFormat.json,
         });
         return {
             response,
@@ -132,7 +132,7 @@ export class LlamaCppTextGenerationModel extends AbstractModel {
     doStreamText(prompt, options) {
         return this.callAPI(prompt, {
             ...options,
-            responseFormat: LlamaCppTextGenerationResponseFormat.deltaIterable,
+            responseFormat: LlamaCppCompletionResponseFormat.deltaIterable,
         });
     }
     extractTextDelta(delta) {
@@ -175,7 +175,7 @@ export class LlamaCppTextGenerationModel extends AbstractModel {
         });
     }
     withSettings(additionalSettings) {
-        return new LlamaCppTextGenerationModel(Object.assign({}, this.settings, additionalSettings));
+        return new LlamaCppCompletionModel(Object.assign({}, this.settings, additionalSettings));
     }
 }
 const llamaCppTextGenerationResponseSchema = z.object({
@@ -200,7 +200,7 @@ const llamaCppTextGenerationResponseSchema = z.object({
         seed: z.number(),
         stop: z.array(z.string()),
         stream: z.boolean(),
-        temp: z.number(),
+        temperature: z.number().optional(), // optional for backwards compatibility
         tfs_z: z.number(),
         top_k: z.number(),
         top_p: z.number(),
@@ -263,7 +263,7 @@ async function createLlamaCppFullDeltaIterableQueue(stream) {
     });
     return queue;
 }
-export const LlamaCppTextGenerationResponseFormat = {
+export const LlamaCppCompletionResponseFormat = {
     /**
      * Returns the response as a JSON object.
      */

package/model-provider/llamacpp/{LlamaCppTextGenerationModel.test.cjs → LlamaCppCompletionModel.test.cjs}

@@ -3,7 +3,7 @@ Object.defineProperty(exports, "__esModule", { value: true });
 const streamText_js_1 = require("../../model-function/generate-text/streamText.cjs");
 const StreamingTestServer_js_1 = require("../../test/StreamingTestServer.cjs");
 const arrayFromAsync_js_1 = require("../../test/arrayFromAsync.cjs");
-const LlamaCppTextGenerationModel_js_1 = require("./LlamaCppTextGenerationModel.cjs");
+const LlamaCppCompletionModel_js_1 = require("./LlamaCppCompletionModel.cjs");
 describe("streamText", () => {
     const server = new StreamingTestServer_js_1.StreamingTestServer("http://127.0.0.1:8080/completion");
     server.setupTestEnvironment();
@@ -17,7 +17,7 @@ describe("streamText", () => {
                 `"mirostat_eta":0.10000000149011612,"mirostat_tau":5.0,"model":"models/llama-2-7b-chat.Q4_K_M.gguf",` +
                 `"n_ctx":4096,"n_keep":0,"n_predict":-1,"n_probs":0,"penalize_nl":true,"penalty_prompt_tokens":[],` +
                 `"presence_penalty":0.0,"repeat_last_n":64,"repeat_penalty":1.100000023841858,"seed":4294967295,` +
-                `"stop":[],"stream":true,"temp":0.800000011920929,"tfs_z":1.0,"top_k":40,"top_p":0.949999988079071,` +
+                `"stop":[],"stream":true,"temperature":0.800000011920929,"tfs_z":1.0,"top_k":40,"top_p":0.949999988079071,` +
                 `"typical_p":1.0,"use_penalty_prompt_tokens":false},"model":"models/llama-2-7b-chat.Q4_K_M.gguf",` +
                 `"prompt":"hello","slot_id":0,"stop":true,"stopped_eos":true,"stopped_limit":false,` +
                 `"stopped_word":false,"stopping_word":"","timings":{"predicted_ms":1054.704,"predicted_n":69,` +
@@ -26,7 +26,7 @@ describe("streamText", () => {
                 `"prompt_per_token_ms":48.845600000000005},"tokens_cached":74,"tokens_evaluated":5,` +
                 `"tokens_predicted":69,"truncated":false}\n\n`,
         ];
-        const stream = await (0, streamText_js_1.streamText)(new LlamaCppTextGenerationModel_js_1.LlamaCppTextGenerationModel().withTextPrompt(), "hello");
+        const stream = await (0, streamText_js_1.streamText)(new LlamaCppCompletionModel_js_1.LlamaCppCompletionModel().withTextPrompt(), "hello");
         // note: space moved to last chunk bc of trimming
         expect(await (0, arrayFromAsync_js_1.arrayFromAsync)(stream)).toStrictEqual([
             "Hello",

package/model-provider/llamacpp/{LlamaCppTextGenerationModel.test.js → LlamaCppCompletionModel.test.js}

@@ -1,7 +1,7 @@
 import { streamText } from "../../model-function/generate-text/streamText.js";
 import { StreamingTestServer } from "../../test/StreamingTestServer.js";
 import { arrayFromAsync } from "../../test/arrayFromAsync.js";
-import { LlamaCppTextGenerationModel } from "./LlamaCppTextGenerationModel.js";
+import { LlamaCppCompletionModel } from "./LlamaCppCompletionModel.js";
 describe("streamText", () => {
     const server = new StreamingTestServer("http://127.0.0.1:8080/completion");
     server.setupTestEnvironment();
@@ -15,7 +15,7 @@ describe("streamText", () => {
                 `"mirostat_eta":0.10000000149011612,"mirostat_tau":5.0,"model":"models/llama-2-7b-chat.Q4_K_M.gguf",` +
                 `"n_ctx":4096,"n_keep":0,"n_predict":-1,"n_probs":0,"penalize_nl":true,"penalty_prompt_tokens":[],` +
                 `"presence_penalty":0.0,"repeat_last_n":64,"repeat_penalty":1.100000023841858,"seed":4294967295,` +
-                `"stop":[],"stream":true,"temp":0.800000011920929,"tfs_z":1.0,"top_k":40,"top_p":0.949999988079071,` +
+                `"stop":[],"stream":true,"temperature":0.800000011920929,"tfs_z":1.0,"top_k":40,"top_p":0.949999988079071,` +
                 `"typical_p":1.0,"use_penalty_prompt_tokens":false},"model":"models/llama-2-7b-chat.Q4_K_M.gguf",` +
                 `"prompt":"hello","slot_id":0,"stop":true,"stopped_eos":true,"stopped_limit":false,` +
                 `"stopped_word":false,"stopping_word":"","timings":{"predicted_ms":1054.704,"predicted_n":69,` +
@@ -24,7 +24,7 @@ describe("streamText", () => {
                 `"prompt_per_token_ms":48.845600000000005},"tokens_cached":74,"tokens_evaluated":5,` +
                 `"tokens_predicted":69,"truncated":false}\n\n`,
         ];
-        const stream = await streamText(new LlamaCppTextGenerationModel().withTextPrompt(), "hello");
+        const stream = await streamText(new LlamaCppCompletionModel().withTextPrompt(), "hello");
         // note: space moved to last chunk bc of trimming
         expect(await arrayFromAsync(stream)).toStrictEqual([
             "Hello",

package/model-provider/llamacpp/LlamaCppFacade.cjs

@@ -3,10 +3,10 @@ Object.defineProperty(exports, "__esModule", { value: true });
 exports.Tokenizer = exports.TextEmbedder = exports.TextGenerator = void 0;
 const LlamaCppApiConfiguration_js_1 = require("./LlamaCppApiConfiguration.cjs");
 const LlamaCppTextEmbeddingModel_js_1 = require("./LlamaCppTextEmbeddingModel.cjs");
-const LlamaCppTextGenerationModel_js_1 = require("./LlamaCppTextGenerationModel.cjs");
+const LlamaCppCompletionModel_js_1 = require("./LlamaCppCompletionModel.cjs");
 const LlamaCppTokenizer_js_1 = require("./LlamaCppTokenizer.cjs");
 function TextGenerator(settings = {}) {
-    return new LlamaCppTextGenerationModel_js_1.LlamaCppTextGenerationModel(settings);
+    return new LlamaCppCompletionModel_js_1.LlamaCppCompletionModel(settings);
 }
 exports.TextGenerator = TextGenerator;
 function TextEmbedder(settings = {}) {

package/model-provider/llamacpp/LlamaCppFacade.d.ts

@@ -1,7 +1,7 @@
 import { ApiConfiguration } from "../../core/api/ApiConfiguration.js";
 import { LlamaCppTextEmbeddingModel, LlamaCppTextEmbeddingModelSettings } from "./LlamaCppTextEmbeddingModel.js";
-import { LlamaCppTextGenerationModel, LlamaCppTextGenerationModelSettings } from "./LlamaCppTextGenerationModel.js";
+import { LlamaCppCompletionModel, LlamaCppCompletionModelSettings } from "./LlamaCppCompletionModel.js";
 import { LlamaCppTokenizer } from "./LlamaCppTokenizer.js";
-export declare function TextGenerator<CONTEXT_WINDOW_SIZE extends number>(settings?: LlamaCppTextGenerationModelSettings<CONTEXT_WINDOW_SIZE>): LlamaCppTextGenerationModel<CONTEXT_WINDOW_SIZE>;
+export declare function TextGenerator<CONTEXT_WINDOW_SIZE extends number>(settings?: LlamaCppCompletionModelSettings<CONTEXT_WINDOW_SIZE>): LlamaCppCompletionModel<CONTEXT_WINDOW_SIZE>;
 export declare function TextEmbedder(settings?: LlamaCppTextEmbeddingModelSettings): LlamaCppTextEmbeddingModel;
 export declare function Tokenizer(api?: ApiConfiguration): LlamaCppTokenizer;

package/model-provider/llamacpp/LlamaCppFacade.js

@@ -1,9 +1,9 @@
 import { LlamaCppApiConfiguration } from "./LlamaCppApiConfiguration.js";
 import { LlamaCppTextEmbeddingModel, } from "./LlamaCppTextEmbeddingModel.js";
-import { LlamaCppTextGenerationModel, } from "./LlamaCppTextGenerationModel.js";
+import { LlamaCppCompletionModel, } from "./LlamaCppCompletionModel.js";
 import { LlamaCppTokenizer } from "./LlamaCppTokenizer.js";
 export function TextGenerator(settings = {}) {
-    return new LlamaCppTextGenerationModel(settings);
+    return new LlamaCppCompletionModel(settings);
 }
 export function TextEmbedder(settings = {}) {
     return new LlamaCppTextEmbeddingModel(settings);

package/model-provider/llamacpp/index.cjs

@@ -33,5 +33,5 @@ var LlamaCppError_js_1 = require("./LlamaCppError.cjs");
 Object.defineProperty(exports, "LlamaCppError", { enumerable: true, get: function () { return LlamaCppError_js_1.LlamaCppError; } });
 exports.llamacpp = __importStar(require("./LlamaCppFacade.cjs"));
 __exportStar(require("./LlamaCppTextEmbeddingModel.cjs"), exports);
-__exportStar(require("./LlamaCppTextGenerationModel.cjs"), exports);
+__exportStar(require("./LlamaCppCompletionModel.cjs"), exports);
 __exportStar(require("./LlamaCppTokenizer.cjs"), exports);