Chore/upgrade llamaindex version (#2440)

* updates to loader to support file upload * adding a todo * upgrade llamaindex * update groq icon * update azure models * update llamaindex version --------- Co-authored-by: Henry <hzj94@hotmail.com>
2026-06-28 19:00:59 +03:00 · 2024-05-22 18:05:08 +05:30
parent e83dcb01b8
commit ff2381741e
22 changed files with 1340 additions and 297 deletions
@@ -1,6 +1,6 @@
 import { ICommonObject, INode, INodeData, INodeOptionsValue, INodeParams } from '../../../src/Interface'
 import { getBaseClasses, getCredentialData, getCredentialParam } from '../../../src/utils'
-import { OpenAI, ALL_AVAILABLE_OPENAI_MODELS } from 'llamaindex'
+import { OpenAI } from 'llamaindex'
 import { getModels, MODEL_TYPE } from '../../../src/modelLoader'

 interface AzureOpenAIConfig {
@@ -10,6 +10,28 @@ interface AzureOpenAIConfig {
    deploymentName?: string
 }

+const ALL_AZURE_OPENAI_CHAT_MODELS = {
+    'gpt-35-turbo': { contextWindow: 4096, openAIModel: 'gpt-3.5-turbo' },
+    'gpt-35-turbo-16k': {
+        contextWindow: 16384,
+        openAIModel: 'gpt-3.5-turbo-16k'
+    },
+    'gpt-4': { contextWindow: 8192, openAIModel: 'gpt-4' },
+    'gpt-4-32k': { contextWindow: 32768, openAIModel: 'gpt-4-32k' },
+    'gpt-4-turbo': {
+        contextWindow: 128000,
+        openAIModel: 'gpt-4-turbo'
+    },
+    'gpt-4-vision-preview': {
+        contextWindow: 128000,
+        openAIModel: 'gpt-4-vision-preview'
+    },
+    'gpt-4-1106-preview': {
+        contextWindow: 128000,
+        openAIModel: 'gpt-4-1106-preview'
+    }
+}
+
 class AzureChatOpenAI_LlamaIndex_ChatModels implements INode {
    label: string
    name: string
@@ -90,7 +112,7 @@ class AzureChatOpenAI_LlamaIndex_ChatModels implements INode {
    }

    async init(nodeData: INodeData, _: string, options: ICommonObject): Promise<any> {
-        const modelName = nodeData.inputs?.modelName as keyof typeof ALL_AVAILABLE_OPENAI_MODELS
+        const modelName = nodeData.inputs?.modelName as keyof typeof ALL_AZURE_OPENAI_CHAT_MODELS
        const temperature = nodeData.inputs?.temperature as string
        const maxTokens = nodeData.inputs?.maxTokens as string
        const topP = nodeData.inputs?.topP as string
@@ -36,7 +36,7 @@ class ChatAnthropic_LlamaIndex_ChatModels implements INode {
            {
                label: 'Model Name',
                name: 'modelName',
-                type: 'options',
+                type: 'asyncOptions',
                loadMethod: 'listModels',
                default: 'claude-3-haiku'
            },
@@ -0,0 +1,100 @@
+import { ICommonObject, INode, INodeData, INodeOptionsValue, INodeParams } from '../../../src/Interface'
+import { MODEL_TYPE, getModels } from '../../../src/modelLoader'
+import { getBaseClasses, getCredentialData, getCredentialParam } from '../../../src/utils'
+import { ALL_AVAILABLE_MISTRAL_MODELS, MistralAI } from 'llamaindex'
+
+class ChatMistral_LlamaIndex_ChatModels implements INode {
+    label: string
+    name: string
+    version: number
+    type: string
+    icon: string
+    category: string
+    description: string
+    tags: string[]
+    baseClasses: string[]
+    credential: INodeParams
+    inputs: INodeParams[]
+
+    constructor() {
+        this.label = 'ChatMistral'
+        this.name = 'chatMistral_LlamaIndex'
+        this.version = 1.0
+        this.type = 'ChatMistral'
+        this.icon = 'MistralAI.svg'
+        this.category = 'Chat Models'
+        this.description = 'Wrapper around ChatMistral LLM specific for LlamaIndex'
+        this.baseClasses = [this.type, 'BaseChatModel_LlamaIndex', ...getBaseClasses(MistralAI)]
+        this.tags = ['LlamaIndex']
+        this.credential = {
+            label: 'Connect Credential',
+            name: 'credential',
+            type: 'credential',
+            credentialNames: ['mistralAIApi']
+        }
+        this.inputs = [
+            {
+                label: 'Model Name',
+                name: 'modelName',
+                type: 'asyncOptions',
+                loadMethod: 'listModels',
+                default: 'mistral-tiny'
+            },
+            {
+                label: 'Temperature',
+                name: 'temperature',
+                type: 'number',
+                step: 0.1,
+                default: 0.9,
+                optional: true
+            },
+            {
+                label: 'Max Tokens',
+                name: 'maxTokensToSample',
+                type: 'number',
+                step: 1,
+                optional: true,
+                additionalParams: true
+            },
+            {
+                label: 'Top P',
+                name: 'topP',
+                type: 'number',
+                step: 0.1,
+                optional: true,
+                additionalParams: true
+            }
+        ]
+    }
+
+    //@ts-ignore
+    loadMethods = {
+        async listModels(): Promise<INodeOptionsValue[]> {
+            return await getModels(MODEL_TYPE.CHAT, 'chatMistral_LlamaIndex')
+        }
+    }
+
+    async init(nodeData: INodeData, _: string, options: ICommonObject): Promise<any> {
+        const temperature = nodeData.inputs?.temperature as string
+        const modelName = nodeData.inputs?.modelName as keyof typeof ALL_AVAILABLE_MISTRAL_MODELS
+        const maxTokensToSample = nodeData.inputs?.maxTokensToSample as string
+        const topP = nodeData.inputs?.topP as string
+
+        const credentialData = await getCredentialData(nodeData.credential ?? '', options)
+        const apiKey = getCredentialParam('mistralAIAPIKey', credentialData, nodeData)
+
+        const obj: Partial<MistralAI> = {
+            temperature: parseFloat(temperature),
+            model: modelName,
+            apiKey: apiKey
+        }
+
+        if (maxTokensToSample) obj.maxTokens = parseInt(maxTokensToSample, 10)
+        if (topP) obj.topP = parseFloat(topP)
+
+        const model = new MistralAI(obj)
+        return model
+    }
+}
+
+module.exports = { nodeClass: ChatMistral_LlamaIndex_ChatModels }
@@ -0,0 +1,221 @@
+import { INode, INodeData, INodeParams } from '../../../src/Interface'
+import { getBaseClasses } from '../../../src/utils'
+import { OllamaParams, Ollama } from 'llamaindex'
+
+class ChatOllama_LlamaIndex_ChatModels implements INode {
+    label: string
+    name: string
+    version: number
+    type: string
+    icon: string
+    category: string
+    description: string
+    tags: string[]
+    baseClasses: string[]
+    inputs: INodeParams[]
+
+    constructor() {
+        this.label = 'ChatOllama'
+        this.name = 'chatOllama_LlamaIndex'
+        this.version = 1.0
+        this.type = 'ChatOllama'
+        this.icon = 'Ollama.svg'
+        this.category = 'Chat Models'
+        this.description = 'Wrapper around ChatOllama LLM specific for LlamaIndex'
+        this.baseClasses = [this.type, 'BaseChatModel_LlamaIndex', ...getBaseClasses(Ollama)]
+        this.tags = ['LlamaIndex']
+        this.inputs = [
+            {
+                label: 'Base URL',
+                name: 'baseUrl',
+                type: 'string',
+                default: 'http://localhost:11434'
+            },
+            {
+                label: 'Model Name',
+                name: 'modelName',
+                type: 'string',
+                placeholder: 'llama3'
+            },
+            {
+                label: 'Temperature',
+                name: 'temperature',
+                type: 'number',
+                description:
+                    'The temperature of the model. Increasing the temperature will make the model answer more creatively. (Default: 0.8). Refer to <a target="_blank" href="https://github.com/jmorganca/ollama/blob/main/docs/modelfile.md#valid-parameters-and-values">docs</a> for more details',
+                step: 0.1,
+                default: 0.9,
+                optional: true
+            },
+            {
+                label: 'Top P',
+                name: 'topP',
+                type: 'number',
+                description:
+                    'Works together with top-k. A higher value (e.g., 0.95) will lead to more diverse text, while a lower value (e.g., 0.5) will generate more focused and conservative text. (Default: 0.9). Refer to <a target="_blank" href="https://github.com/jmorganca/ollama/blob/main/docs/modelfile.md#valid-parameters-and-values">docs</a> for more details',
+                step: 0.1,
+                optional: true,
+                additionalParams: true
+            },
+            {
+                label: 'Top K',
+                name: 'topK',
+                type: 'number',
+                description:
+                    '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). Refer to <a target="_blank" href="https://github.com/jmorganca/ollama/blob/main/docs/modelfile.md#valid-parameters-and-values">docs</a> for more details',
+                step: 1,
+                optional: true,
+                additionalParams: true
+            },
+            {
+                label: 'Mirostat',
+                name: 'mirostat',
+                type: 'number',
+                description:
+                    'Enable Mirostat sampling for controlling perplexity. (default: 0, 0 = disabled, 1 = Mirostat, 2 = Mirostat 2.0). Refer to <a target="_blank" href="https://github.com/jmorganca/ollama/blob/main/docs/modelfile.md#valid-parameters-and-values">docs</a> for more details',
+                step: 1,
+                optional: true,
+                additionalParams: true
+            },
+            {
+                label: 'Mirostat ETA',
+                name: 'mirostatEta',
+                type: 'number',
+                description:
+                    'Influences how quickly the algorithm responds to feedback from the generated text. A lower learning rate will result in slower adjustments, while a higher learning rate will make the algorithm more responsive. (Default: 0.1) Refer to <a target="_blank" href="https://github.com/jmorganca/ollama/blob/main/docs/modelfile.md#valid-parameters-and-values">docs</a> for more details',
+                step: 0.1,
+                optional: true,
+                additionalParams: true
+            },
+            {
+                label: 'Mirostat TAU',
+                name: 'mirostatTau',
+                type: 'number',
+                description:
+                    'Controls the balance between coherence and diversity of the output. A lower value will result in more focused and coherent text. (Default: 5.0) Refer to <a target="_blank" href="https://github.com/jmorganca/ollama/blob/main/docs/modelfile.md#valid-parameters-and-values">docs</a> for more details',
+                step: 0.1,
+                optional: true,
+                additionalParams: true
+            },
+            {
+                label: 'Context Window Size',
+                name: 'numCtx',
+                type: 'number',
+                description:
+                    'Sets the size of the context window used to generate the next token. (Default: 2048) Refer to <a target="_blank" href="https://github.com/jmorganca/ollama/blob/main/docs/modelfile.md#valid-parameters-and-values">docs</a> for more details',
+                step: 1,
+                optional: true,
+                additionalParams: true
+            },
+            {
+                label: 'Number of GPU',
+                name: 'numGpu',
+                type: 'number',
+                description:
+                    'The number of layers to send to the GPU(s). On macOS it defaults to 1 to enable metal support, 0 to disable. Refer to <a target="_blank" href="https://github.com/jmorganca/ollama/blob/main/docs/modelfile.md#valid-parameters-and-values">docs</a> for more details',
+                step: 1,
+                optional: true,
+                additionalParams: true
+            },
+            {
+                label: 'Number of Thread',
+                name: 'numThread',
+                type: 'number',
+                description:
+                    'Sets the number of threads to use during computation. By default, Ollama will detect this for optimal performance. It is recommended to set this value to the number of physical CPU cores your system has (as opposed to the logical number of cores). Refer to <a target="_blank" href="https://github.com/jmorganca/ollama/blob/main/docs/modelfile.md#valid-parameters-and-values">docs</a> for more details',
+                step: 1,
+                optional: true,
+                additionalParams: true
+            },
+            {
+                label: 'Repeat Last N',
+                name: 'repeatLastN',
+                type: 'number',
+                description:
+                    'Sets how far back for the model to look back to prevent repetition. (Default: 64, 0 = disabled, -1 = num_ctx). Refer to <a target="_blank" href="https://github.com/jmorganca/ollama/blob/main/docs/modelfile.md#valid-parameters-and-values">docs</a> for more details',
+                step: 1,
+                optional: true,
+                additionalParams: true
+            },
+            {
+                label: 'Repeat Penalty',
+                name: 'repeatPenalty',
+                type: 'number',
+                description:
+                    'Sets how strongly to penalize repetitions. A higher value (e.g., 1.5) will penalize repetitions more strongly, while a lower value (e.g., 0.9) will be more lenient. (Default: 1.1). Refer to <a target="_blank" href="https://github.com/jmorganca/ollama/blob/main/docs/modelfile.md#valid-parameters-and-values">docs</a> for more details',
+                step: 0.1,
+                optional: true,
+                additionalParams: true
+            },
+            {
+                label: 'Stop Sequence',
+                name: 'stop',
+                type: 'string',
+                rows: 4,
+                placeholder: 'AI assistant:',
+                description:
+                    'Sets the stop sequences to use. Use comma to seperate different sequences. Refer to <a target="_blank" href="https://github.com/jmorganca/ollama/blob/main/docs/modelfile.md#valid-parameters-and-values">docs</a> for more details',
+                optional: true,
+                additionalParams: true
+            },
+            {
+                label: 'Tail Free Sampling',
+                name: 'tfsZ',
+                type: 'number',
+                description:
+                    '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). Refer to <a target="_blank" href="https://github.com/jmorganca/ollama/blob/main/docs/modelfile.md#valid-parameters-and-values">docs</a> for more details',
+                step: 0.1,
+                optional: true,
+                additionalParams: true
+            }
+        ]
+    }
+
+    async init(nodeData: INodeData): Promise<any> {
+        const temperature = nodeData.inputs?.temperature as string
+        const baseUrl = nodeData.inputs?.baseUrl as string
+        const modelName = nodeData.inputs?.modelName as string
+        const topP = nodeData.inputs?.topP as string
+        const topK = nodeData.inputs?.topK as string
+        const mirostat = nodeData.inputs?.mirostat as string
+        const mirostatEta = nodeData.inputs?.mirostatEta as string
+        const mirostatTau = nodeData.inputs?.mirostatTau as string
+        const numCtx = nodeData.inputs?.numCtx as string
+        const numGpu = nodeData.inputs?.numGpu as string
+        const numThread = nodeData.inputs?.numThread as string
+        const repeatLastN = nodeData.inputs?.repeatLastN as string
+        const repeatPenalty = nodeData.inputs?.repeatPenalty as string
+        const stop = nodeData.inputs?.stop as string
+        const tfsZ = nodeData.inputs?.tfsZ as string
+
+        const obj: OllamaParams = {
+            model: modelName,
+            options: {},
+            config: {
+                host: baseUrl
+            }
+        }
+
+        if (temperature) obj.options.temperature = parseFloat(temperature)
+        if (topP) obj.options.top_p = parseFloat(topP)
+        if (topK) obj.options.top_k = parseFloat(topK)
+        if (mirostat) obj.options.mirostat = parseFloat(mirostat)
+        if (mirostatEta) obj.options.mirostat_eta = parseFloat(mirostatEta)
+        if (mirostatTau) obj.options.mirostat_tau = parseFloat(mirostatTau)
+        if (numCtx) obj.options.num_ctx = parseFloat(numCtx)
+        if (numGpu) obj.options.main_gpu = parseFloat(numGpu)
+        if (numThread) obj.options.num_thread = parseFloat(numThread)
+        if (repeatLastN) obj.options.repeat_last_n = parseFloat(repeatLastN)
+        if (repeatPenalty) obj.options.repeat_penalty = parseFloat(repeatPenalty)
+        if (tfsZ) obj.options.tfs_z = parseFloat(tfsZ)
+        if (stop) {
+            const stopSequences = stop.split(',')
+            obj.options.stop = stopSequences
+        }
+
+        const model = new Ollama(obj)
+        return model
+    }
+}
+
+module.exports = { nodeClass: ChatOllama_LlamaIndex_ChatModels }
@@ -1,6 +1,6 @@
 import { ICommonObject, INode, INodeData, INodeOptionsValue, INodeParams } from '../../../src/Interface'
 import { getBaseClasses, getCredentialData, getCredentialParam } from '../../../src/utils'
-import { OpenAI, ALL_AVAILABLE_OPENAI_MODELS } from 'llamaindex'
+import { OpenAI, OpenAISession, ALL_AVAILABLE_OPENAI_MODELS } from 'llamaindex'
 import { getModels, MODEL_TYPE } from '../../../src/modelLoader'

 class ChatOpenAI_LlamaIndex_LLMs implements INode {
@@ -115,8 +115,9 @@ class ChatOpenAI_LlamaIndex_LLMs implements INode {
        if (maxTokens) obj.maxTokens = parseInt(maxTokens, 10)
        if (topP) obj.topP = parseFloat(topP)
        if (timeout) obj.timeout = parseInt(timeout, 10)
+        const openai = new OpenAISession(obj)

-        const model = new OpenAI(obj)
+        const model = new OpenAI({ ...obj, session: openai })
        return model
    }
 }
@@ -0,0 +1,71 @@
+import { ICommonObject, INode, INodeData, INodeParams } from '../../../src/Interface'
+import { getBaseClasses, getCredentialData, getCredentialParam } from '../../../src/utils'
+import { TogetherLLM, OpenAI } from 'llamaindex'
+
+class ChatTogetherAI_LlamaIndex_ChatModels implements INode {
+    label: string
+    name: string
+    version: number
+    type: string
+    icon: string
+    category: string
+    description: string
+    tags: string[]
+    baseClasses: string[]
+    credential: INodeParams
+    inputs: INodeParams[]
+
+    constructor() {
+        this.label = 'ChatTogetherAI'
+        this.name = 'chatTogetherAI_LlamaIndex'
+        this.version = 1.0
+        this.type = 'ChatTogetherAI'
+        this.icon = 'togetherai.png'
+        this.category = 'Chat Models'
+        this.description = 'Wrapper around ChatTogetherAI LLM specific for LlamaIndex'
+        this.baseClasses = [this.type, 'BaseChatModel_LlamaIndex', ...getBaseClasses(TogetherLLM)]
+        this.tags = ['LlamaIndex']
+        this.credential = {
+            label: 'Connect Credential',
+            name: 'credential',
+            type: 'credential',
+            credentialNames: ['togetherAIApi']
+        }
+        this.inputs = [
+            {
+                label: 'Model Name',
+                name: 'modelName',
+                type: 'string',
+                placeholder: 'mixtral-8x7b-32768',
+                description: 'Refer to <a target="_blank" href="https://docs.together.ai/docs/inference-models">models</a> page'
+            },
+            {
+                label: 'Temperature',
+                name: 'temperature',
+                type: 'number',
+                step: 0.1,
+                default: 0.9,
+                optional: true
+            }
+        ]
+    }
+
+    async init(nodeData: INodeData, _: string, options: ICommonObject): Promise<any> {
+        const temperature = nodeData.inputs?.temperature as string
+        const modelName = nodeData.inputs?.modelName as string
+
+        const credentialData = await getCredentialData(nodeData.credential ?? '', options)
+        const togetherAIApiKey = getCredentialParam('togetherAIApiKey', credentialData, nodeData)
+
+        const obj: Partial<OpenAI> = {
+            temperature: parseFloat(temperature),
+            model: modelName,
+            apiKey: togetherAIApiKey
+        }
+
+        const model = new TogetherLLM(obj)
+        return model
+    }
+}
+
+module.exports = { nodeClass: ChatTogetherAI_LlamaIndex_ChatModels }
@@ -0,0 +1,80 @@
+import { ICommonObject, INode, INodeData, INodeOptionsValue, INodeParams } from '../../../src/Interface'
+import { MODEL_TYPE, getModels } from '../../../src/modelLoader'
+import { getBaseClasses, getCredentialData, getCredentialParam } from '../../../src/utils'
+import { Groq, OpenAI } from 'llamaindex'
+
+class ChatGroq_LlamaIndex_ChatModels implements INode {
+    label: string
+    name: string
+    version: number
+    type: string
+    icon: string
+    category: string
+    description: string
+    tags: string[]
+    baseClasses: string[]
+    credential: INodeParams
+    inputs: INodeParams[]
+
+    constructor() {
+        this.label = 'ChatGroq'
+        this.name = 'chatGroq_LlamaIndex'
+        this.version = 1.0
+        this.type = 'ChatGroq'
+        this.icon = 'groq.png'
+        this.category = 'Chat Models'
+        this.description = 'Wrapper around Groq LLM specific for LlamaIndex'
+        this.baseClasses = [this.type, 'BaseChatModel_LlamaIndex', ...getBaseClasses(Groq)]
+        this.tags = ['LlamaIndex']
+        this.credential = {
+            label: 'Connect Credential',
+            name: 'credential',
+            type: 'credential',
+            credentialNames: ['groqApi'],
+            optional: true
+        }
+        this.inputs = [
+            {
+                label: 'Model Name',
+                name: 'modelName',
+                type: 'asyncOptions',
+                loadMethod: 'listModels',
+                placeholder: 'llama3-70b-8192'
+            },
+            {
+                label: 'Temperature',
+                name: 'temperature',
+                type: 'number',
+                step: 0.1,
+                default: 0.9,
+                optional: true
+            }
+        ]
+    }
+
+    //@ts-ignore
+    loadMethods = {
+        async listModels(): Promise<INodeOptionsValue[]> {
+            return await getModels(MODEL_TYPE.CHAT, 'groqChat')
+        }
+    }
+
+    async init(nodeData: INodeData, _: string, options: ICommonObject): Promise<any> {
+        const temperature = nodeData.inputs?.temperature as string
+        const modelName = nodeData.inputs?.modelName as string
+
+        const credentialData = await getCredentialData(nodeData.credential ?? '', options)
+        const groqApiKey = getCredentialParam('groqApiKey', credentialData, nodeData)
+
+        const obj: Partial<OpenAI> = {
+            temperature: parseFloat(temperature),
+            model: modelName,
+            apiKey: groqApiKey
+        }
+
+        const model = new Groq(obj)
+        return model
+    }
+}
+
+module.exports = { nodeClass: ChatGroq_LlamaIndex_ChatModels }