YI

Yi-34Bx2-MoE-60B

YiGGUF by mradermacherUpdated 2024-08-02
Download on Hugging Face
Parameters60.8B
Context195K
Layers60
Architecturellama
Smallest quant12.8 GB
Quick answer

Yi-34Bx2-MoE-60B needs about 39.2 GB of memory at Q4_K_M (a 36.7 GB file) with an 8K context window, so it fits on a 48 GB GPU. This repository has 21 quants, from 12.8 GB (IQ1_S) to 49.9 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 8.1 GB.

Yi-34Bx2-MoE-60B: 60.8B Mixture of Experts LLaMA model in GGUF format

Yi-34Bx2-MoE-60B is a 60.8B parameter Mixture of Experts LLaMA model. It supports 200000 context tokens and is released under the apache-2.0 license. The model is available in multiple GGUF quantizations.

Good at

  • Large context handling
  • Mixture of Experts inference
  • English & Chinese language support

Limits

  • No vision support
  • Requires high VRAM for larger quants
  • License is apache-2.0

Best for

  • Research on large LLMs
  • Fine‑tuning with limited GPU memory
  • Benchmarking Mixture of Experts models

Who it’s for. Researchers and developers needing a large Mixture of Experts LLaMA model for experimentation. The model is suitable for users with access to GPUs capable of handling 60.8B parameters in GGUF format.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S12.8 GB 15.3 GB
Too big Get file →
IQ1_M14.1 GB 16.6 GB
Too big Get file →
IQ2_XXS16.2 GB 18.7 GB
Too big Get file →
IQ2_XS18.0 GB 20.5 GB
Too big Get file →
IQ2_S18.7 GB 21.2 GB
Too big Get file →
IQ2_M20.4 GB 22.9 GB
Too big Get file →
Q2_K22.4 GB 24.9 GB
Too big Get file →
IQ3_XXS23.5 GB 26.0 GB
Too big Get file →
IQ3_XS25.0 GB 27.5 GB
Too big Get file →
Q3_K_S26.3 GB 28.8 GB
Too big Get file →
IQ3_S26.4 GB 28.9 GB
Too big Get file →
IQ3_M27.1 GB 29.6 GB
Too big Get file →
Q3_K_M29.2 GB 31.8 GB
Too big Get file →
Q3_K_L31.8 GB 34.3 GB
Too big Get file →
IQ4_XS32.5 GB 35.0 GB
Too big Get file →
Q4_034.5 GB 37.0 GB
Too big Get file →
Q4_K_S34.6 GB 37.1 GB
Too big Get file →
Q4_K_M36.7 GB 39.2 GB
Too big Get file →
Q5_K_S41.9 GB 44.4 GB
Too big Get file →
Q5_K_M43.1 GB 45.6 GB
Too big Get file →
Q6_K49.9 GB 52.4 GB
Too big Get file →

For 8 GB VRAM, use IQ1_S (12.8 GB file, 15.3 GB memory). For 12 GB VRAM, IQ1_M (14.07 GB file, 16.6 GB memory). For 16 GB VRAM, IQ2_S (18.68 GB file, 21.2 GB memory).

What GPU do you need for Yi-34Bx2-MoE-60B?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: the best quant that fits is IQ1_S (12.8 GB file, about 15.3 GB in use).
  • 24 GB VRAM: the best quant that fits is IQ2_M (20.4 GB file, about 22.9 GB in use).
  • 32 GB VRAM: the best quant that fits is Q3_K_M (29.2 GB file, about 31.8 GB in use).
  • 48 GB VRAM: the best quant that fits is Q5_K_M (43.1 GB file, about 45.6 GB in use).

Which Yi-34Bx2-MoE-60B quant should you choose?

Q4_K_M is the usual starting point for Yi-34Bx2-MoE-60B: it balances file size and quality well. Here it is a 36.7 GB file.

If memory is tight, the smallest file is IQ1_S at 12.8 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters60.8B (approximate, mixture of experts)
Layers60
Trained context200,000 tokens
KV cache (fp16)240 KB per token

Source

Quantized bymradermacher
FamilyYi
Licenseapache-2.0
Downloads (HF)1,247
Last updated2024-08-02

Frequently asked questions about Yi-34Bx2-MoE-60B

How much VRAM does Yi-34Bx2-MoE-60B need?

About 39.2 GB at Q4_K_M with an 8K context window. That is the 36.7 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Yi-34Bx2-MoE-60B on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 15.3 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Yi-34Bx2-MoE-60B GGUF?

It depends on the quant: IQ1_S is 12.8 GB, IQ1_M is 14.1 GB, IQ2_XXS is 16.2 GB, IQ2_XS is 18.0 GB, IQ2_S is 18.7 GB, IQ2_M is 20.4 GB and 15 more.

What context length does Yi-34Bx2-MoE-60B support?

The GGUF header lists a trained context length of 200,000 tokens (195K). The KV cache grows with the context you use: about 2.0 GB at 8K and 8.1 GB at 32K.

Where can I download Yi-34Bx2-MoE-60B GGUF?

The files are on Hugging Face in the mradermacher/Yi-34Bx2-MoE-60B-i1-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Yi-34Bx2-MoE-60B use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run Yi-34Bx2-MoE-60B on 8 GB VRAM?

Use the IQ1_S quant, which requires 15.3 GB memory at 8k context, so 8 GB VRAM is insufficient.

Does the model support vision tasks?

No, the model only supports text and does not include vision capabilities.

What license applies to this model?

The model is released under the apache-2.0 license.

Other GGUFs of Yi-34Bx2-MoE-60B

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Yi models

All Yi →

Other large models

All large models →