AI model developers

Qwen publishes models from 3B to 235B parameters focused on workstation-grade inference.

meta-llama publishes models from 1B to 71B parameters focused on workstation-grade inference.

deepseek-ai publishes models from 3B to 685B parameters focused on workstation-grade inference.

microsoft publishes models from 4B to 14B parameters focused on edge-friendly quantizations.

lmstudio-community publishes models from 4B to 30B parameters focused on workstation-grade inference.

unsloth publishes models from 1B to 32B parameters focused on edge-friendly quantizations.

mistralai publishes models from 7B to 675B parameters focused on high-throughput coders.

google publishes models from 0B to 27B parameters focused on lightweight edge deployment.

openai-community publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

ibm-granite publishes models from 2B to 8B parameters focused on edge-friendly quantizations.

MiniMaxAI publishes models from 7B to 456B parameters focused on workstation-grade inference.

NousResearch publishes models from 8B to 71B parameters focused on workstation-grade inference.

openai publishes models from 20B to 120B parameters focused on workstation-grade inference.

RedHatAI publishes models from 70B to 90B parameters focused on workstation-grade inference.

trl-internal-testing publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

allenai publishes models from 1B to 7B parameters focused on edge-friendly quantizations.

black-forest-labs publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

EleutherAI publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

facebook publishes models from 1B to 7B parameters focused on edge-friendly quantizations.

GSAI-ML publishes models from 8B to 8B parameters focused on edge-friendly quantizations.

HuggingFaceTB publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

moonshotai publishes models from 49B to 1000B parameters focused on workstation-grade inference.

nvidia publishes models from 9B to 71B parameters focused on workstation-grade inference.

tencent publishes models from 1B to 7B parameters focused on edge-friendly quantizations.

zai-org publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

01-ai publishes models from 34B to 34B parameters focused on workstation-grade inference.

ai-forever publishes models from 13B to 13B parameters focused on edge-friendly quantizations.

AI-MO publishes models from 72B to 72B parameters focused on workstation-grade inference.

Alibaba-NLP publishes models from 5B to 5B parameters focused on edge-friendly quantizations.

apple publishes models from 1B to 1B parameters focused on edge-friendly quantizations.

baichuan-inc publishes models from 32B to 32B parameters focused on workstation-grade inference.

bigcode publishes models from 3B to 3B parameters focused on edge-friendly quantizations.

bigscience publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

BSC-LT publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

codellama publishes models from 34B to 34B parameters focused on workstation-grade inference.

context-labs publishes models from 3B to 3B parameters focused on edge-friendly quantizations.

dicta-il publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

distilbert publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

dphn publishes models from 34B to 34B parameters focused on workstation-grade inference.

EssentialAI publishes models from 8B to 8B parameters focused on edge-friendly quantizations.

Gensyn publishes models from 5B to 5B parameters focused on edge-friendly quantizations.

google-bert publishes models from 0B to 0B parameters focused on edge-friendly quantizations.

google-t5 publishes models from 3B to 3B parameters focused on edge-friendly quantizations.

hmellor publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

HuggingFaceH4 publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

HuggingFaceM4 publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

huggyllama publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

ibm-research publishes models from 3B to 3B parameters focused on edge-friendly quantizations.

IlyaGusev publishes models from 8B to 8B parameters focused on edge-friendly quantizations.

inference-net publishes models from 3B to 3B parameters focused on edge-friendly quantizations.

kaitchup publishes models from 4B to 4B parameters focused on edge-friendly quantizations.

LiquidAI publishes models from 2B to 2B parameters focused on edge-friendly quantizations.

liuhaotian publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

llamafactory publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

lmsys publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

mlx-community publishes models from 20B to 20B parameters focused on workstation-grade inference.

nari-labs publishes models from 2B to 2B parameters focused on edge-friendly quantizations.

numind publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

OpenPipe publishes models from 14B to 14B parameters focused on edge-friendly quantizations.

parler-tts publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

petals-team publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

rednote-hilab publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

rinna publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

skt publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

sshleifer publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

swiss-ai publishes models from 8B to 8B parameters focused on edge-friendly quantizations.

TinyLlama publishes models from 1B to 1B parameters focused on edge-friendly quantizations.

Tongyi-MAI publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

vikhyatk publishes models from 7B to 7B parameters focused on edge-friendly quantizations.

WeiboAI publishes models from 2B to 2B parameters focused on edge-friendly quantizations.