Closed sooraj007 closed 1 month ago
Please provide a link to the model you used so I can investigate. From my tests, this model works as expected.
"models:pull": "node-llama-cpp pull --dir ./models \"https://huggingface.co/bartowski/Meta-Llama-3.1-8B-Instruct-GGUF/resolve/main/Meta-Llama-3.1-8B-Instruct-Q5_K_M.gguf\"",
so this is what i am using and i created project using this npm create --yes node-llama-cpp@beta
I haven't done any customization to the code just created the project and try to run it and got those errors , but modal able to run but getting those error kind of messages
I see that the model was updated since I added it to the recommendation list, and it now includes a non-standard chat template that's incompatible with generic Jinja handlers. I'll release a new beta version that uses a different source for Llama 3.1 model recommendations that includes a standard chat template, and seems to work as expected in my tests.
Thanks for reporting this issue :)
:tada: This issue has been resolved in version 3.0.0-beta.40 :tada:
The release is available on:
v3.0.0-beta.40
v3.0.0-beta.40
Your semantic-release bot :package::rocket:
Hey bro you are great, love your commitment, thanks
On Tue, 30 Jul 2024, 11:28 pm github-actions[bot], @.***> wrote:
🎉 This issue has been resolved in version 3.0.0-beta.40 🎉
The release is available on:
- v3.0.0-beta.40
- npm package @.*** dist-tag) https://www.npmjs.com/package/node-llama-cpp/v/3.0.0-beta.40
- GitHub release https://github.com/withcatai/node-llama-cpp/releases/tag/v3.0.0-beta.40
- v3.0.0-beta.40
Your semantic-release https://github.com/semantic-release/semantic-release bot 📦🚀
— Reply to this email directly, view it on GitHub https://github.com/withcatai/node-llama-cpp/issues/275#issuecomment-2258905188, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACYUHIFJWHPS6FI6RDIFXNDZO7H5ZAVCNFSM6AAAAABLUJK7OKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDENJYHEYDKMJYHA . You are receiving this because you authored the thread.Message ID: @.***>
@giladgd Hi again facing the same I pull the latest besta and tested , the same issue is throwing
{ "name": "node_llama", "private": true, "version": "0.0.0", "main": "./dist/index.js", "type": "module", "types": "./dist/index.d.ts", "files": [ "dist/", "package.json", "README.md" ], "exports": { ".": { "import": "./dist/index.js", "node": "./dist/index.js", "types": "./dist/index.d.ts", "default": "./dist/index.js" } }, "engines": { "node": ">=18.0.0" }, "scripts": { "postinstall": "npm run models:pull", "models:pull": "node-llama-cpp pull --dir ./models \"https://huggingface.co/bartowski/Meta-Llama-3.1-8B-Instruct-GGUF/resolve/main/Meta-Llama-3.1-8B-Instruct-Q5_K_M.gguf\"", "start": "vite-node ./src/index.ts", "start:build": "node ./dist/index.ts", "prebuild": "rimraf ./dist ./tsconfig.tsbuildinfo", "build": "tsc --build tsconfig.json --force", "lint": "npm run lint:eslint", "lint:eslint": "eslint --ext .js --ext .ts --report-unused-disable-directives .", "format": "npm run lint:eslint -- --fix", "clean": "rm -rf ./node_modules ./dist ./tsconfig.tsbuildinfo ./models" }, "dependencies": { "chalk": "^5.3.0", "node-llama-cpp": "^3.0.0-beta.40" }, "devDependencies": { "@types/node": "^20.14.2", "@typescript-eslint/eslint-plugin": "^7.12.0", "@typescript-eslint/parser": "^7.12.0", "eslint": "^8.46.0", "eslint-plugin-import": "^2.29.1", "eslint-plugin-jsdoc": "^46.9.0", "eslint-plugin-n": "^17.8.1", "rimraf": "^5.0.7", "tslib": "^2.6.3", "typescript": "^5.4.5", "vite-node": "^1.4.0" } }
G:\personaldev\node-api-boilerplae-creator\node_llama>npm run start
node_llama@0.0.0 start vite-node ./src/index.ts
[node-llama-cpp] Error creating Jinja template chat wrapper. Falling back to resolve other chat wrappers. Error: Error: The provided Jinja template failed that sanity test: Error: Expected iterable type in for loop: got UndefinedValue at JinjaTemplateChatWrapper._runSanityTest (file:///G:/personaldev/node-api-boilerplae-creator/node_llama/node_modules/node-llama-cpp/src/chatWrappers/generic/JinjaTemplateChatWrapper.ts:424:19) at new JinjaTemplateChatWrapper (file:///G:/personaldev/node-api-boilerplae-creator/node_llama/node_modules/node-llama-cpp/src/chatWrappers/generic/JinjaTemplateChatWrapper.ts:128:47) at resolveChatWrapper (file:///G:/personaldev/node-api-boilerplae-creator/node_llama/node_modules/node-llama-cpp/src/chatWrappers/utils/resolveChatWrapper.ts:228:24) at new LlamaChat (file:///G:/personaldev/node-api-boilerplae-creator/node_llama/node_modules/node-llama-cpp/src/evaluator/LlamaChat/LlamaChat.ts:303:17) at new LlamaChatSession (file:///G:/personaldev/node-api-boilerplae-creator/node_llama/node_modules/node-llama-cpp/src/evaluator/LlamaChatSession/LlamaChatSession.ts:303:22) at G:\personaldev\node-api-boilerplae-creator\node_llama\src\index.ts:17:17 at ViteNodeRunner.runModule (file:///G:/personaldev/node-api-boilerplae-creator/node_llama/node_modules/vite-node/dist/client.mjs:362:5) at ViteNodeRunner.directRequest (file:///G:/personaldev/node-api-boilerplae-creator/node_llama/node_modules/vite-node/dist/client.mjs:346:5) at ViteNodeRunner.cachedRequest (file:///G:/personaldev/node-api-boilerplae-creator/node_llama/node_modules/vite-node/dist/client.mjs:189:14) at ViteNodeRunner.executeFile (file:///G:/personaldev/node-api-boilerplae-creator/node_llama/node_modules/vite-node/dist/client.mjs:161:12)
screenshot attached below
@sooraj007 I’ve updated the model recommendations to use a different source for Llama 3.1. Your code still refers to the older model.
Run this command again to generate a new project with the new model:
npm create --yes node-llama-cpp@beta
Issue description
When attempting to run llama 3.1 using the node-llama-cpp latest beta version, an error occurs during the creation of the Jinja template chat wrapper. The specific error message indicates a problem with the sanity test for the provided Jinja template.
Expected Behavior
The Jinja template chat wrapper should be created without errors, allowing llama 3.1 to run successfully.
Actual Behavior
The process fails with an error indicating an issue with the sanity test for the Jinja template.
Error
Error: The provided Jinja template failed that sanity test: Error: Expected iterable type in for loop: got UndefinedValue at JinjaTemplateChatWrapper._runSanityTest (file:///G:/personaldev/node-api-boilerplae-creator/node_llama/node_modules/node-llama-cpp/src/chatWrappers/generic/JinjaTemplateChatWrapper.ts:424:19) at new JinjaTemplateChatWrapper (file:///G:/personaldev/node-api-boilerplae-creator/node_llama/node_modules/node-llama-cpp/src/chatWrappers/generic/JinjaTemplateChatWrapper.ts:128:47) at resolveChatWrapper (file:///G:/personaldev/node-api-boilerplae-creator/node_llama/node_modules/node-llama-cpp/src/chatWrappers/utils/resolveChatWrapper.ts:228:24) at new LlamaChat (file:///G:/personaldev/node-api-boilerplae-creator/node_llama/node_modules/node-llama-cpp/src/evaluator/LlamaChat/LlamaChat.ts:303:17) at new LlamaChatSession (file:///G:/personaldev/node-api-boilerplae-creator/node_llama/node_modules/node-llama-cpp/src/evaluator/LlamaChatSession/LlamaChatSession.ts:303:22) at G:\personaldev\node-api-boilerplae-creator\node_llama\src\index.ts:18:17 at ViteNodeRunner.runModule (file:///G:/personaldev/node-api-boilerplae-creator/node_llama/node_modules/vite-node/dist/client.mjs:362:5) at ViteNodeRunner.directRequest (file:///G:/personaldev/node-api-boilerplae-creator/node_llama/node_modules/vite-node/dist/client.mjs:346:5) at ViteNodeRunner.cachedRequest (file:///G:/personaldev/node-api-boilerplae-creator/node_llama/node_modules/vite-node/dist/client.mjs:189:14) at ViteNodeRunner.executeFile (file:///G:/personaldev/node-api-boilerplae-creator/node_llama/node_modules/vite-node/dist/client.mjs:161:12)
Steps to reproduce
npm run start
Created using the tempalte command provided in the documentaion
My Environment
Node.js Version: 21.x Operating System: windows Version of node-llama-cpp: Latest beta version : 3.0.0-beta.39
Additional Context
Its working correctly but the above error warning is showing
Relevant Features Used
Are you willing to resolve this issue by submitting a Pull Request?
No, I don’t have the time and I’m okay to wait for the community / maintainers to resolve this issue.