Model Finder

Most popular local LLM - runs on consumer hardware with excellent performance

Anthropic's most intelligent model - excels at complex reasoning, coding, and analysis

OpenAI's flagship omni model - text, vision, and audio in one

Best open speech recognition - 99 languages supported

Excellent balance of capability and efficiency for local deployment

Industry-standard image generation model with excellent quality and flexibility

State-of-the-art image generation with exceptional prompt following

Latest Llama with improved tool use and multilingual capabilities

Meta's largest and most capable open model with state-of-the-art performance

The original Mistral - set new standards for 7B models

Google's flagship with 2M token context - can process entire codebases

Small, fast, and cheap GPT-4o for most tasks

State-of-the-art coding model - rivals GPT-4 on coding benchmarks

Classic SD model with massive ecosystem of fine-tunes and LoRAs

State-of-the-art image generation with exceptional quality

Fast FLUX with Apache license for commercial use

OpenAI's latest image generation model

Best-in-class English embedding model for RAG applications

8x faster Whisper with minimal quality loss

Industry-leading voice synthesis with cloning

Highly efficient MoE model - GPT-3.5 quality at fraction of compute

State-of-the-art MoE model rivaling GPT-4 at fraction of compute

Google's latest multimodal model with native tool use and real-time streaming

Best coding model for consumer GPUs

Fast FLUX variant - 4-step generation with Apache 2.0 license

Multi-lingual, multi-functionality, multi-granularity embedding

Classic lightweight embedding - runs on CPU

Efficient multimodal model for consumer hardware

Alibaba's flagship model - rivals GPT-4 on many benchmarks

Best-in-class 7B model with 128K context

Fast and cost-effective Claude for high-volume applications

Fast Gemini with 1M context for high-volume applications

Distilled SDXL for near-real-time generation (1-4 steps)

Highest quality FLUX via API

Efficient embedding model for resource-constrained environments

CTranslate2-optimized Whisper - 4x faster inference

Compact model for edge deployment and low-resource environments

Efficient 12B model with 128K context - great for consumer GPUs

Efficient 9B model with strong instruction following

GPT-4 with vision and 128K context

Strong coding performance in efficient package

Largest open SD3 model with best quality

Highest quality FLUX - API access only

Controlled generation for SDXL - pose, depth, canny, etc.

Google's highest quality image generation

Microsoft's high-quality embedding model

Matryoshka embedding with variable dimension support

Higher quality classic embedding

Best open-source voice cloning with 17 languages

Fast version of SD3.5 Large (4-step generation)

Large multimodal model with image understanding capabilities

Large MoE model with excellent efficiency per active parameter

Sweet spot between capability and efficiency

Remarkably capable 3.8B model for edge deployment

Most capable Claude 3 model for complex tasks

OpenAI's reasoning model - thinks before answering for complex problems

Efficient vision-language model for consumer hardware

DPO-tuned Mistral with excellent chat performance

Mistral's flagship code model

Massive MoE coding model with excellent performance

Latest SD architecture with improved text rendering and composition

Animation module for Stable Diffusion models

Best-in-class text rendering in images

Smallest BGE for edge deployment

State-of-the-art embedding with 128K context

Balanced speed and accuracy for most use cases

Generate music from text descriptions

Strong performance in compact size

Google's largest open model with strong reasoning

State-of-the-art open vision-language model

Efficient agentic model for consumer hardware

Popular fine-tune with great function calling

Compact vision-language model

Efficient MoE coding model for consumer hardware

Compact Code Llama for edge deployment

Aesthetic-focused model optimized for pleasing images

Aesthetic-focused model rivaling Midjourney

Efficient E5 variant

High-quality English embedding with 8K context

Generate speech, music, and sound effects from text

Mistral's flagship model with strong reasoning and function calling

Smallest Gemma 2 for edge deployment

Efficient MoE with excellent code and reasoning

Balanced Claude 3 model for most use cases

Faster, cheaper reasoning model for coding and STEM

NVIDIA's instruction-tuned model optimized for helpfulness

Compact model trained on 3T tokens - surprisingly capable

State-of-the-art open vision-language model

Efficient coding model for consumer GPUs

State-of-the-art open video generation model

LLM-based embedding with 32K context

Task-specific multilingual embeddings

Efficient Whisper for edge deployment

High-quality zero-shot TTS with voice cloning

Unified speech translation across 100+ languages

Compact model for resource-constrained environments

Efficient reasoning model with long context

Optimized for RAG and agentic workflows

Fastest Claude 3 for simple tasks

Best open model for agentic tasks and function calling

Smaller uncensored Dolphin

Efficient LLaVA for consumer GPUs

Tiny vision model - runs anywhere

Compact coding model for resource-constrained environments

Strong coding model with good debugging capability

Strong Code Llama with infilling support

High-quality open video generation with Apache license

Efficient GTE with long context support

Balanced music generation model

Generate variable-length audio up to 47 seconds

Smallest Llama for mobile and embedded deployment

Strong reasoning and math in compact form

Strong bilingual (English/Chinese) model

Efficient RAG and tool-use model

Smallest Gemini for cost-sensitive applications

xAI's flagship model with real-time information access

High-quality 7B chat model trained with C-RLFT

Strong open code model trained on The Stack v2

Popular chat model for consumer GPUs

Uncensored model for research and creative use

Strong OCR and chart understanding

Mistral's math-specialized model

Meta's largest coding model

Efficient Code Llama for consumer hardware

Google's coding model with infilling support

SD 2.1 with improved quality at 768px

Efficient DiT model capable of 4K generation

Strong Chinese-English image generation

Premium embedding API with best-in-class retrieval

Natural-sounding TTS with style control via text descriptions

Fast multilingual TTS for edge deployment

Efficient 10.7B model with depth up-scaling

Efficient chat model

Fast chat model for edge deployment

Intel-optimized chat model

Small but capable model from HuggingFace

Novel Mamba-Transformer hybrid with 256K context

Multi-language coding model trained on The Stack v2

Strong instruction-following coding model

Efficient 4K image generation with small footprint

Efficient Parler TTS for edge deployment

Efficient music generation for consumer hardware

Edit speech with natural voice cloning

Smallest Qwen for embedded and mobile

Efficient bilingual model

Strong Chinese-English model from Shanghai AI Lab

RLHF-tuned model with high MT-Bench scores

Efficient code model for consumer hardware

Strong chat model from LMSYS

IBM's enterprise-grade code model

Smallest Qwen coder for embedded/mobile

Efficient StarCoder for consumer hardware

Multilingual image generation with strong Russian support

Tencent's bilingual DiT model

Smallest Whisper for real-time on CPU

Smaller Bark for faster generation

Generate sound effects and ambient audio from text

Novel TTS with fine-grained prosody control

Microsoft's reasoning-focused model

Stability AI's flagship chat model

Efficient IBM code model

Smaller Jamba with same 256K context

Tiny CodeGemma for embedded systems

Efficient WizardCoder for consumer hardware

Compact bilingual model for edge

Fast Grok for simple tasks

Efficient Chinese-English model

Microsoft's MoE model for complex reasoning

Smallest StarCoder2 for edge deployment

Smaller Orca for edge deployment

Smallest StableLM for mobile/edge

Tiny model for embedded systems

Amazon's general-purpose LLM via Bedrock

Tiny coding model for embedded systems

Compact StarCoder for edge deployment

Efficient model with 65K context support

Linear attention model with infinite context potential

TII's largest open model

Strong Chinese chat model

MosaicML's chat model with 8K context

Largest open multilingual model - 46 languages

Strong open multilingual model

First commercially usable instruction-tuned model

Open reproduction of LLaMA trained on RedPajama

Instruction-tuned BLOOM

Fully open reproduction of LLaMA

Research model with full training checkpoints

Smaller OpenLLaMA variant

Smaller Pythia variant

Compute-optimal GPT trained by Cerebras