PH

cognitivecomputations Dolphin 2.9.1 Phi 3 Kensho 4.5B GGUF smashed

PhiGGUF by PrunaAIUpdated 2026-03-25
Download on Hugging Face
Parameters6.54B
Context4K
Layers56
Architecturellama
Smallest quant2.4 GB
Quick answer

cognitivecomputations Dolphin 2.9.1 Phi 3 Kensho 4.5B GGUF smashed needs about 10.1 GB of memory at Q4_K_M (a 4.0 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 14 quants, from 2.4 GB (Q2_K) to 6.9 GB (Q8_0).

cognitivecomputations Dolphin 2.9.1 Phi 3 Kensho 4.5B GGUF smashed is a 6.54B-parameter Phi model in GGUF format, quantized by PrunaAI. Its trained context length is 4K tokens. The repository has 14 quant files, from 2.4 GB (Q2_K) to 6.9 GB (Q8_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K2.4 GB 5.8 GB
Fits Get file →
Q3_K_S2.9 GB 6.2 GB
Fits Get file →
Q3_K_M3.2 GB 6.5 GB
Fits Get file →
Q3_K_L3.5 GB 6.8 GB
Fits Get file →
Q4_03.7 GB 7.0 GB
Fits Get file →
Q4_K_S3.7 GB 7.1 GB
Fits Get file →
Q4_K_M4.0 GB 7.3 GB
Tight Get file →
Q4_14.1 GB 7.4 GB
Tight Get file →
Q5_04.5 GB 7.8 GB
Tight Get file →
Q5_K_S4.5 GB 7.8 GB
Tight Get file →
Q5_K_M4.6 GB 8.0 GB
Tight Get file →
Q5_14.9 GB 8.2 GB
Too big Get file →
Q6_K5.4 GB 8.7 GB
Too big Get file →
Q8_06.9 GB 10.3 GB
Too big Get file →

What GPU do you need for cognitivecomputations Dolphin 2.9.1 Phi 3 Kensho 4.5B GGUF smashed?

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: the best quant that fits is Q6_K (5.4 GB file, about 11.5 GB in use).
  • 16 GB VRAM: the best quant that fits is Q8_0 (6.9 GB file, about 13.1 GB in use).

Which cognitivecomputations Dolphin 2.9.1 Phi 3 Kensho 4.5B GGUF smashed quant should you choose?

Q4_K_M is the usual starting point for cognitivecomputations Dolphin 2.9.1 Phi 3 Kensho 4.5B GGUF smashed: it balances file size and quality well. Here it is a 4.0 GB file.

The Q8_0 file is 6.9 GB, about 1.8 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is Q2_K at 2.4 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
Parameters6.54B
Layers56
Trained context4,096 tokens
KV cache (fp16)672 KB per token

Source

Quantized byPrunaAI
FamilyPhi
Downloads (HF)363
Last updated2026-03-25

Frequently asked questions about cognitivecomputations Dolphin 2.9.1 Phi 3 Kensho 4.5B GGUF smashed

How much VRAM does cognitivecomputations Dolphin 2.9.1 Phi 3 Kensho 4.5B GGUF smashed need?

About 10.1 GB at Q4_K_M with an 8K context window. That is the 4.0 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 cognitivecomputations Dolphin 2.9.1 Phi 3 Kensho 4.5B GGUF smashed on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 8.6 GB, so you would need a larger GPU or CPU offloading.

What is the file size of cognitivecomputations Dolphin 2.9.1 Phi 3 Kensho 4.5B GGUF smashed GGUF?

It depends on the quant: Q2_K is 2.4 GB, Q3_K_S is 2.9 GB, Q3_K_M is 3.2 GB, Q3_K_L is 3.5 GB, Q4_0 is 3.7 GB, Q4_K_S is 3.7 GB and 8 more.

What context length does cognitivecomputations Dolphin 2.9.1 Phi 3 Kensho 4.5B GGUF smashed support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 5.6 GB at 8K and 2.8 GB at 4K.

Where can I download cognitivecomputations Dolphin 2.9.1 Phi 3 Kensho 4.5B GGUF smashed GGUF?

The files are on Hugging Face in the PrunaAI/cognitivecomputations-Dolphin-2.9.1-Phi-3-Kensho-4.5B-GGUF-smashed repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Similar models

Browse all →

More from PrunaAI

All by PrunaAI →

More Phi models

All Phi →

Other small models

All small models →