Apple
Mac Studio M3 Ultra 96GB with 96 GB unified memory. Third-generation Apple Silicon built on 3nm process with dynamic caching GPU architecture, significantly improving AI inference efficiency.
Architecture
Apple M3 is built on TSMC's 3nm process, the first consumer chips at this node. It introduces Dynamic Caching for more efficient GPU memory allocation and hardware-accelerated ray tracing.
AI Relevance
Dynamic Caching improves GPU utilization for compute workloads including ML inference. The M3 Ultra with up to 512 GB unified memory can theoretically hold even unquantized 70B models, though memory bandwidth remains the throughput bottleneck.
M3's dynamic caching GPU architecture allocates local memory in hardware in real-time, improving GPU utilization for AI workloads. The M3 Max reaches 400 GB/s bandwidth, competitive with mid-range discrete GPUs.
Agentic Coding
CThis model is still usable for agentic-coding, but it is not the most specialized pick. It belongs to a current frontier family for local AI. It should run, but memory headroom will be limited. Known channels: huggingface, ollama, lm-studio.
Chat
CThis model is a direct match for chat. It belongs to a current frontier family for local AI. It fits natively with comfortable headroom. Known channels: huggingface, ollama, lm-studio.
Coding
CThis model is a direct match for coding. It belongs to a current frontier family for local AI. It should run, but memory headroom will be limited. Known channels: huggingface, ollama, lm-studio.
RAG
CThis model is a direct match for rag. It sits in the middle of the current model mix. It fits natively with comfortable headroom. Known channels: huggingface, ollama, lm-studio.
Reasoning
CThis model is a direct match for reasoning. It belongs to a current frontier family for local AI. It fits natively with comfortable headroom. Known channels: huggingface, ollama, lm-studio.
Qwen3-Coder 30B A3B InstructQwen3-VL 30B A3B InstructQwen3-Coder-Next
Qwen3.5 35B A3BQwen 2.5 Coder 32BQwen3.5 27Bgemma 3 27b it
Devstral Small 2 24B InstructDevstral Small 1.1Codestral 2 25.08
Llama 3.3 70BQwen3.5 9BQwen 2.5 72BQwen 2.5 VL 72BDeepSeek R1 0528 Qwen3 8BQwen3.5 4BQwen2.5 3B Instruct
gemma 2bQwen2.5 1.5B Instruct
DeepSeek R1 671BDevstral 2 123B Instruct
GLM-5
Kimi K2.5Mistral Large 3Mistral Small 4 119BQwen3-Coder 480B A35B InstructQwen3.5 122B A10BDeepSeek V3 671BMixtral 8x22BQwen 3 235B A22BQwen3.5 397B A17BLlama 4 Maverick 17B 128E
Command A 111BJust out of reach
High-quality models that need a bit more memory
Upgrade paths
See what you unlock with more powerful hardware
Upgrade options
NVIDIA A100 80GBNext step up~$15,000 MSRP
~$3,999 MSRP
~$8,000 MSRP
