Arcee Trinity Nano 6B
Trinity · Apache 2.0
Mixture of Experts: 6B total parameters, 1B active per token.
Arcee AI's smallest Mixture-of-Experts model. 6B total parameters with only 1B active per token (128 experts, 8 selected + 1 shared). Runs on consumer GPUs, edge devices, and mobile; fits at full Q8_0 precision on any GPU with 8 GB of VRAM. Still a preview release, so expect rough edges in some workflows. US-built, Apache 2.0 licensed.
- Parameters
- 6B
- Architecture
- MoE (1B active)
- Context
- 131,072 tokens
- Released
- 2026-01-27
- Engines
- llama.cpp, ollama, LM Studio
- Builder Tools
- Ollama, LM Studio, Open WebUI
Parameters
6B
VRAM
5.4 GB
Context
128K
Formats
4
GPUs
43
Arcee Trinity Nano 6B (6B) requires 5.4 GB VRAM at recommended quality (Q5_K_M). At efficient quality (Q4_K_M), it fits in 4.8 GB VRAM, making it compatible with the NVIDIA RTX 4060 Laptop (40-60W). On NVIDIA Grace Blackwell Ultra GB300, expect approximately 1411 tok/s at Q8_0. For the best experience, Starter AI Desktop ($582) is recommended.
Source: OwnRig methodology
5.4 GB
Q5_K_M
4.41 GB
128K tokens
Chat
VRAM Requirements
| Quality | Quantization | VRAM | File Size |
|---|---|---|---|
| full | Q8_0 | 7.5 GB | 6.53 GB |
| recommended | Q5_K_M | 5.4 GB | 4.41 GB |
| efficient | Q4_K_M | 4.8 GB | 3.79 GB |
| compressed | Q3_K_M | 3.9 GB | 2.91 GB |
Context Length Impact
KV cache VRAM at Q5_K_M quality. Longer context = more memory.
| Context | KV Cache | Total VRAM |
|---|---|---|
| 2K | 0 MB | 5.4 GB |
| 4K | 102 MB | 5.5 GB |
| 8K | 102 MB | 5.5 GB |
| 16K | 307 MB | 5.7 GB |
| 32K | 614 MB | 6 GB |
| 64K | 1.2 GB | 6.6 GB |
| 128K | 2.3 GB | 7.7 GB |
Compatible GPUs
43 devicesShowing 43 of 43 entries
Builder Context
Arcee Trinity Nano 6B is commonly used with Ollama, LM Studio, Open WebUI. For an AI coding workflow, pair it with an embedding model like nomic-embed-text for local RAG.
Frequently Asked Questions
- How much VRAM does Arcee Trinity Nano 6B need?
- Arcee Trinity Nano 6B requires 5.4 GB VRAM at recommended quality (Q5_K_M). At lower quality settings, it can fit in as little as 3.9 GB.
- What is the best GPU for Arcee Trinity Nano 6B?
- The NVIDIA Grace Blackwell Ultra GB300 delivers the best performance for Arcee Trinity Nano 6B, achieving 1411 tok/s at Q8_0 with an excellent rating.
- Can I run Arcee Trinity Nano 6B on an RTX 4060 Ti?
- Yes. On the NVIDIA GeForce RTX 4060 Ti 16GB, Arcee Trinity Nano 6B runs at 51 tok/s (Q8_0, excellent).
- What quantization should I use for Arcee Trinity Nano 6B?
- For the best quality, use Q5_K_M (5.4 GB VRAM). If your GPU has limited VRAM, Q3_K_M (3.9 GB) is the most efficient option with acceptable quality.
- Is Arcee Trinity Nano 6B good for coding?
- Yes. Arcee Trinity Nano 6B is used with Ollama, LM Studio, Open WebUI for local AI coding. For the best coding experience, pair it with an embedding model for local RAG.
Data confidence: estimated. Source
VRAM requirements are calculated from model parameters and may vary by inference engine, context length, and batch size. Performance estimates are based on community benchmarks and should be verified for your specific configuration.Trinity is a trademark of its respective owner. OwnRig is not affiliated with or endorsed by the model creator.