CA

Carnice MoE 35B A3B APEX

CarniceGGUF by mudlerUpdated 2026-08-17
Download on Hugging Face
Parameters34.7B
Context256K
Layers40
Architectureqwen35moe
Smallest quant14.3 GB
Quick answer

Carnice MoE 35B A3B APEX needs about 24.0 GB of memory at its main file (a 22.8 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 8 quants, from 14.3 GB to 69.4 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 2.7 GB.

Carnice MoE 35B A3B APEX – a 35B Mixture‑of‑Experts model fine‑tuned for agentic workflows

Carnice MoE 35B A3B APEX is a 35.7B‑parameter Qwen3.5‑MoE model with 3B active parameters per token. It is trained on real Hermes Agent execution traces and optimized for tool‑calling and agentic tasks. The model is released under the apache‑2.0 license.

Good at

  • Tool‑calling with Hermes Agent
  • Agentic workflow execution
  • Multi‑step reasoning

Limits

  • Requires large VRAM for 8k context
  • No vision support
  • Only supports text
  • License is apache‑2.0

Best for

  • Building autonomous agents
  • Command‑line tool integration
  • Browser‑assisted workflows

Who it’s for. Developers building agentic applications that need robust tool‑calling and multi‑step reasoning. The model is suitable for users who can allocate sufficient GPU memory for MoE inference.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—14.3 GB 15.5 GB
Too big Get file →
—17.3 GB 18.5 GB
Too big Get file →
—17.3 GB 18.5 GB
Too big Get file →
—22.8 GB 24.0 GB
Too big Get file →
—22.8 GB 24.0 GB
Too big Get file →
—25.6 GB 26.8 GB
Too big Get file →
—25.6 GB 26.8 GB
Too big Get file →
F1669.4 GB 70.5 GB
Too big Get file →

For 8k context, the 69.38 GB F16 quant requires about 70.5 GB VRAM. Smaller quants are not listed, so use the F16 file if you have the memory. If you have less than 70 GB, the model will not run at full 8k context.

What GPU do you need for Carnice MoE 35B A3B APEX?

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 the main file (14.3 GB file, about 15.5 GB in use).
  • 24 GB VRAM: the best quant that fits is the main file (22.8 GB file, about 24.0 GB in use).
  • 32 GB VRAM: the best quant that fits is the main file (25.6 GB file, about 26.8 GB in use).
  • 48 GB VRAM: the best quant that fits is the main file (25.6 GB file, about 26.8 GB in use).

Which Carnice MoE 35B A3B APEX quant should you choose?

If memory is tight, the smallest file is the first one in the table at 14.3 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen35moe
Parameters34.7B (approximate, mixture of experts)
Active parameters3B
Layers40
Trained context262,144 tokens
KV cache (fp16)80 KB per token

Source

Quantized bymudler
FamilyCarnice
Licenseapache-2.0
Downloads (HF)2,520
Last updated2026-08-17

Frequently asked questions about Carnice MoE 35B A3B APEX

How much VRAM does Carnice MoE 35B A3B APEX need?

About 24.0 GB at its main file with an 8K context window. That is the 22.8 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 Carnice MoE 35B A3B APEX on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 15.5 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Carnice MoE 35B A3B APEX GGUF?

It depends on the quant: file is 14.3 GB, file is 17.3 GB, file is 17.3 GB, file is 22.8 GB, file is 22.8 GB, file is 25.6 GB and 2 more.

What context length does Carnice MoE 35B A3B APEX support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.7 GB at 8K and 2.7 GB at 32K.

Where can I download Carnice MoE 35B A3B APEX GGUF?

The files are on Hugging Face in the mudler/Carnice-MoE-35B-A3B-APEX-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 Carnice MoE 35B A3B APEX 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 Carnice MoE 35B A3B APEX on 8 GB VRAM?

No. The smallest available quant is 69.38 GB and requires about 70.5 GB VRAM at 8k context.

Does this model support vision?

No. The model only handles text.

What license applies to this model?

The model is released under the apache‑2.0 license.

Other GGUFs of Carnice-MoE-35B-A3B

Similar models

Browse all →

More from mudler

All by mudler →

More Carnice models

All Carnice →

Other large models

All large models →