Skip to content
@DeltaVML

DeltaVML

DeltaV Machine Learning DeltaVML

DeltaVML is a collection of open source software to facilitate the exploration of local language models for machine learning experiments. Goals include AI assistants for all aspects of digital lifestyle enhancements. Ideals include DAIO (Decentralised Autonomous Intelligent Organization), AGLM (a general learning model), AIOS (Automated Intelligent Operating System) and integrating intelligence into deltaverse. It is the opinion of DeltaVML that intelligence is intelligence. Labelling intelligence as artificial, natural or cosmological is contradictory to intelligence. Intelligence "is" or "is not". Machine learning is one of the components of intelligence.

One working model is provided by koboldai. First inception of a local language model was using GPT4ALL and the alpaca model provided by Stanford university. Kobaldcpp provides my first experience finding a model that runs on both the graphics card and the cpu. DeltaVML believes in a decentralized modular open source future for local langauge models.

The following is the output from a build using Ubuntu 23 on an i7 Desktop with 12 GB of ram and an Nvidia GeForce GTX 970 representing average to below average North American consumer grade hardware in 2023. Replies currently take 2-3 minutes. Parameter settings and fine tuning is being done to increase the performance.

~/koboldcpp$ python koboldcpp.py --threads 8 --blasthreads 8 --useclblast 0 0 --highpriority --contextsize 4096 --blasbatchsize 64 --smartcontext --gpulayers 8 --debugmode --model ggml-gpt4all-j-v1.3-groovy.bin

Welcome to KoboldCpp - Version 1.29 Setting process to Higher Priority - Use Caution High Priority for Linux Set: 0 to 1 Attempting to use CLBlast library for faster prompt ingestion. A compatible clblast will be required. Initializing dynamic library: koboldcpp_clblast.so

Loading model: /home/codephreak/koboldcpp/ggml-gpt4all-j-v1.3-groovy.bin [Threads: 8, BlasThreads: 8, SmartContext: True]


Identified as GPT-J model: (ver 102) Attempting to Load...

System Info: AVX = 0 | AVX2 = 0 | AVX512 = 0 | AVX512_VBMI = 0 | AVX512_VNNI = 0 | FMA = 0 | NEON = 0 | ARM_FMA = 0 | F16C = 0 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 1 | SSE3 = 1 | VSX = 0 |

gptj_v2_model_load: loading model from '/home/codephreak/koboldcpp/ggml-gpt4all-j-v1.3-groovy.bin' - please wait ...

gptj_v2_model_load: n_vocab = 50400 gptj_v2_model_load: n_ctx = 2048 gptj_v2_model_load: n_embd = 4096 gptj_v2_model_load: n_head = 16 gptj_v2_model_load: n_layer = 28 gptj_v2_model_load: n_rot = 64 gptj_v2_model_load: ftype = 2 gptj_v2_model_load: qntvr = 0 gptj_v2_model_load: ggml ctx size = 4505.52 MB

Initializing LEGACY CLBlast (First Run)... Attempting to use: Platform=0, Device=0 (If invalid, program will crash)

Using Platform: NVIDIA CUDA Device: NVIDIA GeForce GTX 970

gptj_v2_model_load: memory_size = 896.00 MB, n_mem = 57344 gptj_v2_model_load: ................................... done gptj_v2_model_load: model size = 3609.38 MB / num tensors = 285

Load Model OK: True Embedded Kobold Lite loaded. Starting Kobold HTTP Server on port 5001 Please connect to custom endpoint at http://localhost:5001


Who is the greek god hermes?

Output:

The Greek God Hermes was the messenger of the gods, responsible for guiding souls to the afterlife after they died. He was depicted with wings, carrying a caduceus staff, and wearing a winged hat. In mythology, he was often associated with trade and commerce, and his name was given to various professions such as merchants, traders, and messengers.

Time Taken - Processing:45.0s (529ms/T), Generation:124.8s (1580ms/T), Total:169.8s

Pinned Loading

  1. awesome-production-machine-learning awesome-production-machine-learning Public

    Forked from EthicalML/awesome-production-machine-learning

    A curated list of awesome open source libraries to deploy, monitor, version and scale your machine learning

    1

  2. ml.ml ml.ml Public

    DeltaV Machine Learning

  3. Open-Assistant Open-Assistant Public

    Forked from LAION-AI/Open-Assistant

    OpenAssistant is a chat-based assistant that understands tasks, can interact with third-party systems, and retrieve information dynamically to do so.

    Python

  4. ChatGLM-6B ChatGLM-6B Public

    Forked from THUDM/ChatGLM-6B

    ChatGLM-6B:开源双语对话语言模型 | An Open Bilingual Dialogue Language Model

    Python 1

  5. privateGPT privateGPT Public

    Forked from zylon-ai/private-gpt

    Interact privately with your documents using the power of GPT, 100% privately, no data leaks

    Python 2

  6. thrust thrust Public

    Forked from NVIDIA/thrust

    The C++ parallel algorithms library.

    C++

Repositories

Showing 10 of 267 repositories
  • cube Public Forked from Professor-Codephreak/cube

    Roblox Foundation Model for 3D Intelligence

    DeltaVML/cube’s past year of commit activity
    Jupyter Notebook 0 47 0 0 Updated Mar 22, 2025
  • vertex-ai-gemini Public Forked from augml/vertex-ai-gemini

    Notebooks, code samples, sample apps, and other resources that demonstrate how to use, develop and manage machine learning and generative AI workflows using Google Cloud Vertex AI.

    DeltaVML/vertex-ai-gemini’s past year of commit activity
    Jupyter Notebook 0 Apache-2.0 95 0 0 Updated Mar 21, 2025
  • Hunyuan3D-2 Public Forked from Tencent/Hunyuan3D-2

    High-Resolution 3D Assets Generation with Large Scale Hunyuan3D Diffusion Models

    DeltaVML/Hunyuan3D-2’s past year of commit activity
    Python 1 686 0 0 Updated Mar 21, 2025
  • modular-account Public Forked from alchemyplatform/modular-account

    Alchemy's Modular Account is a maximally modular, upgradeable smart contract account that is compatible with ERC-4337 and ERC-6900.

    DeltaVML/modular-account’s past year of commit activity
    Solidity 0 GPL-3.0 45 0 0 Updated Mar 4, 2025
  • dkn-compute-node Public Forked from firstbatchxyz/dkn-compute-node

    Compute Node of Dria Knowledge Network.

    DeltaVML/dkn-compute-node’s past year of commit activity
    Rust 0 Apache-2.0 49 0 0 Updated Feb 6, 2025
  • applied-ai-engineering-gemini Public Forked from augml/applied-ai-engineering-samples

    This repository compiles code samples and notebooks demonstrating how to use Generative AI on Google Cloud Vertex AI.

    DeltaVML/applied-ai-engineering-gemini’s past year of commit activity
    Jupyter Notebook 0 Apache-2.0 190 0 0 Updated Dec 19, 2024
  • portkey-gateway Public Forked from augml/portkey-gateway

    A Blazing Fast AI Gateway with integrated Guardrails. Route to 200+ LLMs, 50+ AI Guardrails with 1 fast & friendly API.

    DeltaVML/portkey-gateway’s past year of commit activity
    TypeScript 0 MIT 570 0 0 Updated Dec 6, 2024
  • litellm Public Forked from All-Hands-AI/litellm

    Python SDK, Proxy Server (LLM Gateway) to call 100+ LLM APIs in OpenAI format - [Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthropic, Sagemaker, HuggingFace, Replicate, Groq]

    DeltaVML/litellm’s past year of commit activity
    Python 0 2,596 0 0 Updated Oct 29, 2024
  • textual Public Forked from Textualize/textual

    The lean application framework for Python. Build sophisticated user interfaces with a simple Python API. Run your apps in the terminal and a web browser.

    DeltaVML/textual’s past year of commit activity
    Python 0 MIT 878 0 0 Updated Sep 28, 2024
  • hyperspace-node Public Forked from hyperspaceai/hyperspace-node

    The world’s first decentralized AI network for Mistral7B-based model

    DeltaVML/hyperspace-node’s past year of commit activity
    0 29 0 0 Updated Sep 16, 2024

People

This organization has no public members. You must be a member to see who’s a part of this organization.

Top languages

Loading…

Most used topics

Loading…