---

license: other

license_name: nvidia-open-model-license

license_link: https://developer.download.nvidia.com/licenses/nvidia-open-model-license-agreement-june-2024.pdf

base_model: nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16

tags:

- gguf

- turboquant

- kv-cache-quantization

- nemotron

- nvidia

- mamba2

- hybrid

- llama-cpp

- quantized

library_name: gguf

pipeline_tag: text-generation

---

> [!TIP]

> KV-cache quantization without any fork (recommended, 2026): upstream

> llama.cpp/Ollama now cover this natively — use -ctk q8_0 -ctv q8_0

> (~half KV memory, negligible quality loss: perplexity +0.002–0.05) or

> -ctk q4_0 -ctv q4_0 (~quarter memory, ≈7.6% perplexity increase). In

> Ollama: OLLAMA_KV_CACHE_TYPE=q8_0 with OLLAMA_FLASH_ATTENTION=1. Keep

> K and V types symmetric to stay on the fast fused Flash-Attention path.

> Since April 2026, mainline llama.cpp also applies Hadamard rotation to

> KV activations (PR #21038),

> which greatly improves low-bit KV quality (opt-out:

> LLAMA_ATTN_ROT_DISABLE=1).

>

> The RotorQuant/TurboQuant fork flow below is experimental/legacy: the

> TurboQuant llama.cpp PR was closed without merging (June 2026) and the fork

> is unmaintained relative to mainline. It is NOT required to use this model.

<!-- kv-upstream-note -->

Nemotron-3-Nano-4B-TurboQuant-GGUF-Q4_K_M

GGUF Q4_K_M weight-quantized variant of nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16 optimised for use with TurboQuant KV cache compression via a dedicated llama.cpp fork.

> Important: TurboQuant KV cache types (planar3, iso3) are not available in upstream llama.cpp, standard Ollama, or LM Studio.

> They require a specific llama.cpp fork.

> The GGUF file itself is a standard GGUF and works with any llama.cpp-compatible runtime using normal KV cache types (f16, q8_0, q4_0, etc.).

Hardware compatibility

| Device | VRAM / RAM | Recommendation |

| --- | --- | --- |

| CPU host with ≥8 GB RAM | ~2.4 GB | works via llama.cpp; slower than GPU but no accelerator required |

| Apple Silicon (Metal) | ~2.6 GB | llama.cpp Metal backend; fast on M-series unified memory |

| NVIDIA GPU (partial offload) | split between GPU + RAM | offload as many layers as VRAM allows; rest on CPU |

Overview

This model combines two independent compression techniques:

| Technique | What it does | Requirement |

|-----------|-------------|-------------|

| GGUF Q4_K_M weight quantization | Reduces model size from ~8 GB (BF16) to ~2.2 GB | Any llama.cpp-compatible runtime |

| TurboQuant KV cache compression — random rotation + Lloyd-Max scalar quantization (--cache-type-k planar3 --cache-type-v planar3) | Block-diagonal rotations / random rotation for compressed KV cache | llama-cpp-turboquant fork only |

Quickstart

Option A — RotorQuant KV cache (experimental fork — not required)

You must build from the TurboQuant-enabled llama.cpp fork:

# Clone and build the fork
git clone https://github.com/johndpope/llama-cpp-turboquant.git
cd llama-cpp-turboquant && git checkout feature/planarquant-kv-cache

# CUDA (Windows/Linux)
cmake -B build -DGGML_CUDA=ON -DCMAKE_BUILD_TYPE=Release && cmake --build build -j

# Metal (Apple Silicon)
cmake -B build -DGGML_METAL=ON -DGGML_METAL_EMBED_LIBRARY=ON -DCMAKE_BUILD_TYPE=Release && cmake --build build -j

# Run with TurboQuant KV cache
./build/bin/llama-cli -m Nemotron-3-Nano-4B-TurboQuant-GGUF-Q4_K_M.gguf \
  --cache-type-k planar3 --cache-type-v planar3 \
  -ngl 99 -fa \
  -p "Explain quantum computing"

# Or run as a server
./build/bin/llama-server -m Nemotron-3-Nano-4B-TurboQuant-GGUF-Q4_K_M.gguf \
  --cache-type-k planar3 --cache-type-v planar3 \
  -ngl 99 -fa --jinja

Option B — With standard llama.cpp / LM Studio / Ollama

The GGUF works as a normal quantised model. You won't get TurboQuant-specific KV cache benefits, but standard KV cache quantization (q8_0, q4_0) still reduces VRAM significantly.

llama.cpp (upstream)

llama-cli -m Nemotron-3-Nano-4B-TurboQuant-GGUF-Q4_K_M.gguf \
  --cache-type-k q8_0 --cache-type-v q8_0 \
  -ngl 99 -fa \
  -p "Explain quantum computing"

LM Studio

1. Download the GGUF file and load in LM Studio.
2. Enable Developer Mode (Settings → Developer).
3. In the model loader's advanced settings, set Flash Attention to ON.
4. Set K Cache Quantization and V Cache Quantization to q8_0 (or q4_0 for more aggressive VRAM savings).
5. Note: LM Studio does not currently support TurboQuant's planar3 cache types. Track this feature request for updates.

Ollama

# Standard Ollama does not support TurboQuant cache types.
# Use with default or q8_0 KV cache via OLLAMA_KV_CACHE_TYPE=q8_0
OLLAMA_KV_CACHE_TYPE=q8_0 OLLAMA_FLASH_ATTENTION=1 ollama run majentik/Nemotron-3-Nano-4B-TurboQuant-GGUF-Q4_K_M

Specifications

| Property | Value |

|----------|-------|

| Base Model | nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16 |

| Architecture | Mamba-2 + Transformer hybrid (dense) |

| Parameters | 4B (dense hybrid) |

| Context Length | 262K |

| Weight Quantization | GGUF Q4_K_M (popular 4-bit, best quality/size tradeoff) |

| Original Size (BF16) | ~8 GB |

| Quantized File Size | ~2.2 GB |

| KV Cache (TurboQuant) | 3-bit via --cache-type-k planar3 --cache-type-v planar3 (fork only) |

| KV Cache (standard) | q8_0, q4_0, f16, etc. (any llama.cpp runtime) |

| License | other |

| Modalities | Text only |

| Compatible Runtimes | llama.cpp, LM Studio, Ollama, koboldcpp |

About the RotorQuant / TurboQuant labels

RotorQuant and TurboQuant are this project's release labels, not distinct

quantization algorithms — for any given tier, both brand repos carry

byte-identical weights produced with the standard MLX / llama.cpp quantizers.

No brand-specific speedup is claimed or measured. The KV-cache fork these

labels originally referred to is legacy; for KV-cache memory savings use the

upstream options described above (-ctk/-ctv q8_0, OLLAMA_KV_CACHE_TYPE).

Current Status of TurboQuant in the Ecosystem

| Runtime | TurboQuant Support | Standard KV Quant |

|---------|---------------------|-------------------|

| llama.cpp (upstream) | ❌ Not merged | ✅ q8_0, q4_0, q4_1, iq4_nl, q5_0, q5_1 |

| llama-cpp-turboquant fork | ✅ planar3 | ✅ All standard types |

| LM Studio | ❌ Requested | ✅ Via advanced settings |

| Ollama | ❌ Not supported | ✅ Via OLLAMA_KV_CACHE_TYPE |

| koboldcpp | ❌ Not supported | ✅ Standard types |

Recommended Settings

For VRAM-constrained setups, standard q8_0 KV cache quantization already halves KV cache memory with negligible quality impact. Flash Attention should always be enabled — it is required for V cache quantization and improves memory efficiency regardless.

| VRAM | Suggested Configuration |

|------|------------------------|

| 24 GB (RTX 4090) | Q4_K_M + q8_0 KV cache + Flash Attention, 8K–16K context |

| 16 GB | Q4_K_M + q4_0 KV cache + Flash Attention, 4K–8K context |

| 48+ GB | Q4_K_M + f16 KV cache, full 32K+ context |

See Also

• RotorQuant GitHub
• llama-cpp-turboquant fork
• TurboQuant llama.cpp discussion
• TurboQuant paper (arXiv: 2504.19874)
• Base model: nvidia/NVIDIA-Nemotron-3-Nano-4B-BF16
• Nemotron-3-Nano-4B announcement

Quant trade-off (GGUF lane)

| Quant | Approx size | Use case | Recommendation |

|---|---|---|---|

| Q2_K | ~2.2 GB | Lossy, low-RAM CPU/edge | Resource-constrained inference |

| Q3_K_M | ~2.4 GB | Smaller-than-Q4, modest quality drop | Edge devices with ~16 GB RAM |

| IQ4_XS | ~2.1 GB | Importance-quant 4-bit, smaller than Q4_K_M | Best size/quality at 4-bit |

| Q4_K_M | ~3.0 GB | Balanced default | Recommended for most users |

| Q5_K_M | ~3.1 GB | Higher fidelity than Q4 | Quality-sensitive applications |

| Q6_K | ~3.6 GB | Approaching FP16 quality | High-fidelity CPU/edge |

| Q8_0 | ~4.1 GB | Near-lossless reference | Fidelity-critical work |

| MXFP4_MOE | ~2.2 GB | Microscaling FP4 (MoE-aware) | vLLM / transformers users |

(Current variant — Q4_K_M — is bolded.)

Variants in this family

(Showing 13 sibling variants under majentik/nemotron3-nano-4b-*. The current variant — TurboQuant-GGUF-Q4_K_M — is bolded.)

| Variant | Runtime | Approx size | Use case |

|---|---|---|---|

| TurboQuant-GGUF-Q4_K_M | llama.cpp | ~4.4 GB | Balanced default |