LocalClaw/LLM/Trinity Large Preview (70B MoE)
Open-weight MoE

Trinity Large Preview (70B MoE)

Arcee AI's massive MoE open model. ~400B total parameters, 70B active per forward pass. Ranks near the top of global usage leaderboards. Exceptional versatility across reasoning, coding and chat. Free and open-source. Apache 2.0.

64 GB workstation 48 GB RAM Q4_K_M Enterprise AI
Run with LocalClaw Compare all models
Parameters
70B (MoE, ~400B total)
Minimum RAM
48 GB
Model size
45 GB
Quantization
Q4_K_M

Can Trinity Large Preview (70B MoE) run locally?

Trinity Large Preview (70B MoE) is best for 64 GB workstations and larger Apple Silicon or NVIDIA setups.

Search for trinity-large-preview in LM Studio or another GGUF-compatible runtime.

arcee-ai/Trinity-Large-Preview-GGUF
chatcodereasoningpowerqualitygeneral

Install path

01
Check RAM fitMinimum 48 GB RAM. Start with the Q4_K_M quant.
02
Load the modelSearch trinity-large-preview in LM Studio.
03
Control locallyUse LocalClaw to manage models, agents, chat, channels and scheduled OpenClaw work.

Strengths

  • Ranks #2 globally with 114B monthly tokens
  • Exceptional versatility
  • Apache 2.0
  • Free and open-source

Limitations

  • Requires 48GB+ RAM
  • New model
  • MoE complexity

Best use cases

  • Enterprise AI
  • Complex reasoning
  • Coding
  • Research

Capability profile

speed
3
quality
10
coding
10
reasoning
10

Technical notes

Developer
Arcee AI
License
Apache 2.0
Context window
131,072 tokens
Architecture
Mixture of Experts (MoE), 70B

This model fits these next steps

Hardware fit is based on LocalClaw's RAM tier, model size and quantization metadata. Always leave memory headroom for your OS and runtime.

Workstation fitMac Studio M4 Max 64GBCUDA workstation classNVIDIA GB10 / DGX SparkLarge local models64GB RAM guide

Similar models to compare

DeepSeek V3.2 (37B/671B MoE) 37B (671B MoE)Llama 3.3 (70B) 70BQwen 2.5 (72B) 72B

Where to go next