MI

MiniCPM V 4.6 Thinking Heretic MTP

MiniCPMGGUF by MuXodiousUpdated 2026-05-17
Download on Hugging Face
Parameters773M
Context256K
Layers25
Architectureqwen35
Smallest quant0.6 GB
Quick answer

MiniCPM V 4.6 Thinking Heretic MTP needs about 1.8 GB of memory at Q8_0 (a 0.8 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 3 quants, from 0.6 GB (Q6_K) to 1.6 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 1.7 GB.

MiniCPM V 4.6 Thinking Heretic MTP – a 0.773 B parameter MLLM with chain‑of‑thought reasoning

MiniCPM V 4.6 Thinking Heretic MTP is a 0.773 B parameter model based on the qwen35 architecture. It supports reasoning and vision tasks and uses a 262144‑token context window. The model is released under the apache‑2.0 license.

Good at

  • Chain‑of‑thought reasoning
  • Multimodal vision understanding
  • Long‑context processing

Limits

  • No vision support in standard GGUF without MTP arguments
  • Requires 8 k context for memory estimates
  • Experimental MTP implementation

Best for

  • Complex reasoning tasks
  • Image and video understanding on mobile
  • Multimodal instruction following

Who it’s for. Developers building multimodal applications that need explicit reasoning traces. Researchers testing experimental MTP layers on a lightweight LLM.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q6_K0.6 GB 1.6 GB
Fits Get file →
Q8_00.8 GB 1.8 GB
Fits Get file →
BF161.6 GB 2.5 GB
Fits Get file →

For 8 GB VRAM, use the Q6_K quant (0.65 GB file, ~1.6 GB memory at 8 k context). If 12 GB VRAM is available, Q8_0 (0.83 GB, ~1.8 GB memory) offers higher precision. BF16 (1.56 GB, ~2.5 GB memory) is suitable for 16 GB VRAM setups.

What GPU do you need for MiniCPM V 4.6 Thinking Heretic MTP?

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: the best quant that fits is BF16 (1.6 GB file, about 2.5 GB in use).

Which MiniCPM V 4.6 Thinking Heretic MTP quant should you choose?

Q8_0 (0.8 GB) is the middle option in this repository. Close to lossless. It costs roughly twice the memory of a 4-bit file, so it suits cases where you have plenty of VRAM.

If memory is tight, the smallest file is Q6_K at 0.6 GB, with the quality trade-offs that come with fewer bits. Very close to the original quality for most uses, at a clearly larger size than 4-bit.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen35
Parameters773M
Layers25
Trained context262,144 tokens
KV cache (fp16)50 KB per token

Source

Quantized byMuXodious
FamilyMiniCPM
Licenseapache-2.0
Downloads (HF)2,289
Last updated2026-05-17

Frequently asked questions about MiniCPM V 4.6 Thinking Heretic MTP

How much VRAM does MiniCPM V 4.6 Thinking Heretic MTP need?

About 1.8 GB at Q8_0 with an 8K context window. That is the 0.8 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 MiniCPM V 4.6 Thinking Heretic MTP on an 8 GB GPU?

Yes, with BF16: the 1.6 GB file needs about 2.5 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of MiniCPM V 4.6 Thinking Heretic MTP GGUF?

It depends on the quant: Q6_K is 0.6 GB, Q8_0 is 0.8 GB, BF16 is 1.6 GB.

What context length does MiniCPM V 4.6 Thinking Heretic MTP support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.4 GB at 8K and 1.7 GB at 32K.

Where can I download MiniCPM V 4.6 Thinking Heretic MTP GGUF?

The files are on Hugging Face in the MuXodious/MiniCPM-V-4.6-Thinking-heretic-MTP-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 MiniCPM V 4.6 Thinking Heretic MTP 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 this model on 8 GB VRAM?

Yes, the Q6_K quant uses about 1.6 GB of VRAM at 8 k context, which fits within an 8 GB budget.

Does this model support vision tasks?

Yes, it includes vision capabilities but requires the MTP arguments to enable them in llama.cpp.

What license applies to this model?

The model is released under the apache‑2.0 license.

Similar models

Browse all →

More from MuXodious

All by MuXodious →

More MiniCPM models

All MiniCPM →

Other tiny models

All tiny models →