Model Matrix
Every workstation ships with 10+ LLM models pre-loaded via Ollama. Browse by task type or workstation tier to find the right fit.
12 models shown
| Model | Task Type | VRAM | Tier | Use Case Examples |
|---|---|---|---|---|
Llama 3.1 8B Meta | GeneralChatLogic | 6 GB | Pro |
|
Llama 3.1 70B Meta | GeneralLogicReasoning | 48 GB | MAX |
|
Mistral 7B Mistral AI | GeneralChatCreative | 5 GB | Pro |
|
Mixtral 8x7B Mistral AI | GeneralLogicCreative | 26 GB | MAX |
|
CodeLlama 13B Meta | CodingLogic | 10 GB | Pro |
|
DeepSeek Coder 33B DeepSeek | CodingLogic | 20 GB | MAX |
|
Phi-3 Mini Microsoft | LogicReasoningGeneral | 3 GB | Pro |
|
Gemma 2 9B | GeneralChatReasoning | 7 GB | Pro |
|
Qwen 2.5 72B Alibaba | GeneralCodingCreativeLogic | 48 GB | MAX |
|
LLaVA 1.6 13B LLaVA Team | VisionGeneralReasoning | 10 GB | Pro |
|
Nous Hermes 2 Nous Research | CreativeReasoningChat | 8 GB | Pro |
|
Neural Chat 7B Intel | ChatGeneral | 5 GB | Pro |
|
Llama 3.1 8B
Meta
Fast, capable general-purpose model. Excellent balance of speed and quality for everyday tasks.
Use Cases
Llama 3.1 70B
Meta
Meta's flagship large model. Near-GPT-4 quality reasoning and instruction following.
Use Cases
Mistral 7B
Mistral AI
Highly efficient open-source model. Punches above its weight in reasoning and generation.
Use Cases
Mixtral 8x7B
Mistral AI
Mixture-of-experts architecture delivering GPT-3.5 class performance with great efficiency.
Use Cases
CodeLlama 13B
Meta
Specialized for code generation, debugging, and explanation across 20+ programming languages.
Use Cases
DeepSeek Coder 33B
DeepSeek
State-of-the-art coding model rivaling GPT-4 on benchmarks. Exceptional at complex algorithms.
Use Cases
Phi-3 Mini
Microsoft
Small but mighty. Microsoft's most capable compact model, ideal for on-device tasks.
Use Cases
Gemma 2 9B
Google's open-weights model with strong instruction following and safety-tuning.
Use Cases
Qwen 2.5 72B
Alibaba
Alibaba's top-tier multilingual model with outstanding coding and reasoning scores.
Use Cases
LLaVA 1.6 13B
LLaVA Team
Multimodal model that understands images and text. Describe, analyze, and reason over visuals.
Use Cases
Nous Hermes 2
Nous Research
Fine-tuned for complex instruction following, creative writing, and agentic tasks.
Use Cases
Neural Chat 7B
Intel
Optimized by Intel for conversational use cases. Runs efficiently on Intel hardware.
Use Cases