Instructions to use IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF:Q8_0
# Run inference directly in the terminal:
llama cli -hf IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF:Q8_0
# Run inference directly in the terminal:
llama cli -hf IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF:Q8_0

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF:Q8_0

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF:Q8_0

Use Docker

docker model run hf.co/IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF:Q8_0

LM Studio
Jan

vLLM

How to use IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF:Q8_0

Ollama
How to use IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF with Ollama:
```
ollama run hf.co/IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF:Q8_0
```
Unsloth Desktop

How to use IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF:Q8_0

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF:Q8_0"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF with Docker Model Runner:
```
docker model run hf.co/IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF:Q8_0
```

Lemonade

How to use IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF:Q8_0

Run and chat with the model

lemonade run user.Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF-Q8_0

List all available models

lemonade list

Hermes Agent

How to use IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF:Q8_0

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF:Q8_0

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF:Q8_0

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "IsValorum/Nex-N2.5-mini-APEX-I-MiniPlus-V1-GGUF:Q8_0" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

IsValorum commited on 22 days ago

Commit

17a98fa

verified ·

1 Parent(s): 7ccc2fb

Upload README.md with huggingface_hub

Browse files

Files changed (1) hide show

README.md +6 -6

README.md CHANGED Viewed

@@ -33,7 +33,7 @@ Most existing community quantizations are generated by automated bots that apply
 ## <a id="quick-navigation"></a>⚡ Quick Navigation Index
 - [📦 Model Files & Specifications](#model-specifications)
-- [🔬 Comparative Quantization Analysis (vs. Flat Quants & Mudler APEX)](#comparative-analysis)
 - [👁️ Bundled Q8_0 High-Precision Vision Projector](#vision-projector)
 - [💻 Everyday Laptop Benchmarks (23–26+ tok/s on DDR4)](#laptop-benchmarks)
 - [🔥 The 24GB Miracle: Full 256K Context Runs In VRAM!](#context-scaling)
@@ -57,17 +57,17 @@ Most existing community quantizations are generated by automated bots that apply
 ---
 <a id="comparative-analysis"></a>
-## 🔬 Comparative Quantization Analysis (vs. Flat Quants & Mudler APEX)
-The table below breaks down the surgical choices in **APEX-I-MiniPlus** compared directly against standard community automated quants (such as flat bartowski releases) and Mudler's standard APEX-I-Mini recipe:
-| Architectural Component | Flat Community Quants (e.g., bartowski `Q3_K_S` / `IQ3_S`) | Mudler APEX-I-Mini (Standard Recipe) | Our Handcrafted APEX-I-MiniPlus (IsValorum) | Perceived Quality & Real-World Impact |
 | :--- | :--- | :--- | :--- | :--- |
 | **Output Head (`output.weight`)** | Flat **`IQ3_S` / `Q3_K_S`** (~3.44 BPW) | **`Q4_K`** or profile default (~4.5 BPW) | **`Q6_K`** (~6.56 BPW uncompromised) | **Eliminates Syntax & Vocabulary Hallucinations:** Low-bit output heads cause tokenizer classification noise, breaking code indentation, brackets (`{}`, `[]`), math symbols, and domain terms. `Q6_K` preserves near-FP16 output classification. |
 | **Expert Routers (`ffn_gate_inp.weight`)** | Blindly quantized to 3-bit / unoptimized | Tier-quantized or standard linear/Q4 | **`F32` uncompressed** (32.0 BPW, 2 MB/layer) | **Zero Router Drift:** In 256 micro-expert models, even minuscule quantization errors in router logits misdirect tokens to wrong experts. Retaining uncompressed `F32` guarantees 100% routing fidelity with virtually zero memory overhead (~80 MB total). |
 | **Shared Foundation Expert (`ffn_*_shexp`)** | Flat **`IQ3_S` / `Q3_K_S`** (3.44 BPW) | **`Q3_K` / `IQ3_S`** (3.44 BPW) | **`IQ4_NL`** (4.50 BPW non-linear codebook) | **Foundational Knowledge Armor:** The shared expert executes for 100% of tokens. Crushing it to 3 bits degrades common-sense and domain reasoning. `IQ4_NL` maintains high representational fidelity on the universal pathway. |
-| **Core MoE Layers (Middle: 10–29)** | Flat **`IQ3_S` / `Q3_K_S`** (uniform bit-rate across all layers) | Aggressive **`IQ2_S` (2.5 BPW)** | **`IQ3_XXS` (3.06 BPW) + calibrated `imatrix`** | **Above the Quality Threshold:** Mudler's 2-bit `IQ2_S` drops below the critical quality floor for 35B MoEs, resulting in perplexity spikes on reasoning tasks. Our `IQ3_XXS` with imatrix achieves deep compression (272 MiB → 98 MiB per block) without sacrificing logic. |
-| **Edge MoE Layers (Layers 0–9 & 30–39)** | Flat **`IQ3_S` / `Q3_K_S`** (no layer-wise gradient) | `IQ3_S` (limited to first/last 5 layers only) | **`IQ3_S` (expanded to 10 input & 10 output layers)** | **Protected Ingestion & Synthesis:** Half of the model's layers (10 at input, 10 at output) form an armored envelope, preventing initial prompt misunderstanding and final token degeneration during multi-turn generation. |
 | **Attention Gates (`attn_gate.weight`)** | Blindly compressed to 3-bit | Unoptimized / tier default | **`Q8_0`** (8.50 BPW) | **Attention Head Stability:** Attention gates modulate query-key routing across hybrid attention layers. Keeping them in 8-bit prevents attention crosstalk and hallucination over long contexts. |
 | **Attention Output & QKV (`attn_output`, `attn_qkv`)** | Flat **`IQ3_S` / `Q3_K_S`** | Tier default | **`Q6_K` for `attn_output`**, **`IQ3_S` for `attn_qkv`** | **Contextual Retrieval Precision:** Preserves high dynamic range in self-attention projections, ensuring flawless needle-in-a-haystack retrieval across deep 128k–256k context windows. |
 | **Multimodal Vision (`mmproj`)** | Often omitted, or left as uncompressed **`FP16` (~900 MB)** | Often omitted or separate uncompressed `FP16` | **Bundled `Q8_0` (582 MB)** with **27 critical F32/F16 fallbacks** | **Saves ~320 MB VRAM with Zero Loss:** Handcrafted quantization preserves normalization and bias tensors in F32/F16, ensuring razor-sharp OCR, DOM viewport reading, and coordinate detection without visual noise. |

 ## <a id="quick-navigation"></a>⚡ Quick Navigation Index
 - [📦 Model Files & Specifications](#model-specifications)
+- [🔬 Comparative Quantization Analysis (vs. Flat Quants & Generic APEX)](#comparative-analysis)
 - [👁️ Bundled Q8_0 High-Precision Vision Projector](#vision-projector)
 - [💻 Everyday Laptop Benchmarks (23–26+ tok/s on DDR4)](#laptop-benchmarks)
 - [🔥 The 24GB Miracle: Full 256K Context Runs In VRAM!](#context-scaling)
 ---
 <a id="comparative-analysis"></a>
+## 🔬 Comparative Quantization Analysis (vs. Flat Quants & Generic APEX)
+The table below breaks down the surgical choices in **APEX-I-MiniPlus** compared directly against generic automated community quants (uniform flat recipes) and generic APEX baseline recipes:
+| Architectural Component | Generic Automated Quants (Flat `Q3_K_S` / `IQ3_S`) | Generic APEX-I-Mini (Baseline Recipe) | Our Handcrafted APEX-I-MiniPlus (IsValorum) | Perceived Quality & Real-World Impact |
 | :--- | :--- | :--- | :--- | :--- |
 | **Output Head (`output.weight`)** | Flat **`IQ3_S` / `Q3_K_S`** (~3.44 BPW) | **`Q4_K`** or profile default (~4.5 BPW) | **`Q6_K`** (~6.56 BPW uncompromised) | **Eliminates Syntax & Vocabulary Hallucinations:** Low-bit output heads cause tokenizer classification noise, breaking code indentation, brackets (`{}`, `[]`), math symbols, and domain terms. `Q6_K` preserves near-FP16 output classification. |
 | **Expert Routers (`ffn_gate_inp.weight`)** | Blindly quantized to 3-bit / unoptimized | Tier-quantized or standard linear/Q4 | **`F32` uncompressed** (32.0 BPW, 2 MB/layer) | **Zero Router Drift:** In 256 micro-expert models, even minuscule quantization errors in router logits misdirect tokens to wrong experts. Retaining uncompressed `F32` guarantees 100% routing fidelity with virtually zero memory overhead (~80 MB total). |
 | **Shared Foundation Expert (`ffn_*_shexp`)** | Flat **`IQ3_S` / `Q3_K_S`** (3.44 BPW) | **`Q3_K` / `IQ3_S`** (3.44 BPW) | **`IQ4_NL`** (4.50 BPW non-linear codebook) | **Foundational Knowledge Armor:** The shared expert executes for 100% of tokens. Crushing it to 3 bits degrades common-sense and domain reasoning. `IQ4_NL` maintains high representational fidelity on the universal pathway. |
+| **Core MoE Layers (Middle: 10–29)** | Flat **`IQ3_S` / `Q3_K_S`** (uniform bit-rate across all layers) | Aggressive **`IQ2_S` (2.5 BPW)** | **`IQ3_XXS` (3.06 BPW) + calibrated `imatrix`** | **Above the Quality Threshold:** Generic 2-bit `IQ2_S` baselines drop below the critical quality floor for 35B MoEs, resulting in perplexity spikes on reasoning tasks. Our `IQ3_XXS` with imatrix achieves deep compression (272 MiB → 98 MiB per block) without sacrificing logic. |
+| **Edge MoE Layers (Layers 0–9 & 30–39)** | Flat **`IQ3_S` / `Q3_K_S`** (no layer-wise gradient) | `IQ3_S` (limited to first/last 5 layers only in generic recipes) | **`IQ3_S` (expanded to 10 input & 10 output layers)** | **Protected Ingestion & Synthesis:** Half of the model's layers (10 at input, 10 at output) form an armored envelope, preventing initial prompt misunderstanding and final token degeneration during multi-turn generation. |
 | **Attention Gates (`attn_gate.weight`)** | Blindly compressed to 3-bit | Unoptimized / tier default | **`Q8_0`** (8.50 BPW) | **Attention Head Stability:** Attention gates modulate query-key routing across hybrid attention layers. Keeping them in 8-bit prevents attention crosstalk and hallucination over long contexts. |
 | **Attention Output & QKV (`attn_output`, `attn_qkv`)** | Flat **`IQ3_S` / `Q3_K_S`** | Tier default | **`Q6_K` for `attn_output`**, **`IQ3_S` for `attn_qkv`** | **Contextual Retrieval Precision:** Preserves high dynamic range in self-attention projections, ensuring flawless needle-in-a-haystack retrieval across deep 128k–256k context windows. |
 | **Multimodal Vision (`mmproj`)** | Often omitted, or left as uncompressed **`FP16` (~900 MB)** | Often omitted or separate uncompressed `FP16` | **Bundled `Q8_0` (582 MB)** with **27 critical F32/F16 fallbacks** | **Saves ~320 MB VRAM with Zero Loss:** Handcrafted quantization preserves normalization and bias tensors in F32/F16, ensuring razor-sharp OCR, DOM viewport reading, and coordinate detection without visual noise. |