salka/inference-x
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

Compare commits

base: salka:master
Branches Tags
salka:master
salka:v1.0.0
..
compare: salka:v1.0.0
Branches Tags
salka:master
salka:v1.0.0

No commits in common. "master" and "v1.0.0" have entirely different histories.
master ... v1.0.0

22 changed files with 220 additions and 147 deletions
Show Stats Expand all files Collapse all files

3
.github/FUNDING.yml vendored
View File

@ -1,5 +1,6 @@
# Inference-X — Universal Inference Protocol # Inference-X — Universal Inference Protocol
# Free for individuals, researchers, and small teams. # Free for individuals, researchers, and small teams.
# Your support funds development and server infrastructure. # Your support funds development, servers, and solar inference research.
github: ElmadaniS
custom: ["https://paypal.me/ELMADANISALKA"] custom: ["https://paypal.me/ELMADANISALKA"]

2
CONTRIBUTORS.md
View File

@ -2,7 +2,7 @@
## Creator & Lead Developer ## Creator & Lead Developer
- **Salka Elmadani** — Architecture, implementation, and all original code - **Salka Elmadani** — Architecture, implementation, and all original code
- GitHub: [@ElmadaniS](https://git.inference-x.com/salka) - GitHub: [@ElmadaniS](https://github.com/ElmadaniS)
- Email: Elmadani.SALKA@proton.me - Email: Elmadani.SALKA@proton.me
## Infrastructure Partners ## Infrastructure Partners

2
NOTICE
View File

@ -15,7 +15,7 @@ AUTHOR
Location: Morocco Location: Morocco
Contact: Elmadani.SALKA@proton.me Contact: Elmadani.SALKA@proton.me
Website: https://inference-x.com Website: https://inference-x.com
Repository: https://git.inference-x.com/salka/inference-x Repository: https://github.com/ElmadaniS/inference-x
Origin: Morocco 🇲🇦 Origin: Morocco 🇲🇦
──────────────────────────────────────────────────────────────── ────────────────────────────────────────────────────────────────

302
README.md
View File

@ -1,196 +1,262 @@
# Inference-X # Inference-X
[![Build](https://github.com/ElmadaniS/inference-x/actions/workflows/build.yml/badge.svg)](https://github.com/ElmadaniS/inference-x/actions/workflows/build.yml)
[![Release](https://img.shields.io/github/v/release/ElmadaniS/inference-x)](https://github.com/ElmadaniS/inference-x/releases)
[![License](https://img.shields.io/badge/license-BSL--1.1-blue)](LICENSE) [![License](https://img.shields.io/badge/license-BSL--1.1-blue)](LICENSE)
[![Binary Size](https://img.shields.io/badge/binary-305%20KB-brightgreen)](TECHNOLOGY.md) [![Binary Size](https://img.shields.io/badge/binary-305%20KB-brightgreen)](TECHNOLOGY.md)
[![Backends](https://img.shields.io/badge/backends-19-orange)](ARCHITECTURE.md) [![Backends](https://img.shields.io/badge/backends-19-orange)](ARCHITECTURE.md)
**Run AI on your own computer. Private. Free. No internet.** **Better output from the same model.**
Inference-X is a tiny file (305 KB) that lets any computer run AI models locally. It works on old laptops, phones, Raspberry Pi, and datacenters — same file, no setup. Your questions stay on your machine. Nobody sees them. One binary routes any AI model to any hardware — from a microcontroller to a datacenter. Fused computation, adaptive precision, surgical expert loading. No dependencies. No framework. No vendor lock-in.
**[Website](https://inference-x.com)** · **[How it works](TECHNOLOGY.md)** · **[Benchmarks](BENCHMARKS.md)** · **[Vision](VISION.md)** · **[Sponsor](https://git.inference-x.com/salka)** 305 KB. 19 hardware backends. Any model. Any scale.
--- Built in Morocco by [Salka Elmadani](https://x.com/ElmadaniSa13111).
## Start in 30 seconds > *In the Anti-Atlas, our ancestors built khettaras — underground water channels that deliver pure water to villages without pumps, without electricity, without filtration. The water arrives cleaner than any treated supply because the path itself is the filter. Inference-X works the same way: the shortest path produces the cleanest signal.*
```bash **[Website](https://inference-x.com)** · **[How it works](TECHNOLOGY.md)** · **[Benchmarks](BENCHMARKS.md)** · **[Vision](VISION.md)** · **[Sponsor](https://github.com/sponsors/ElmadaniS)**
git clone https://git.inference-x.com/salka/inference-x
cd inference-x && make
./inference-x model.gguf
```
That's it. Download a `.gguf` model from [HuggingFace](https://huggingface.co/models?sort=trending&search=gguf), run the command, talk to AI. No account. No API key. No internet.
Add `--serve 8080` to get a web interface at `localhost:8080`.
---
## What can your computer run?
| Your RAM | Models you can run | What it can do |
|---|---|---|
| **2 GB** | SmolLM2 135M | Simple assistant, quick answers |
| **4 GB** | Phi-3 Mini 3.8B, Llama 3.2 3B | Smart conversations, code help, translations |
| **8 GB** | Mistral 7B, Llama 3.1 8B | Creative writing, analysis, reasoning |
| **16 GB** | DeepSeek R1 14B | Advanced reasoning, expert-level answers |
| **32 GB** | Qwen 2.5 32B | Professional-grade AI |
| **64 GB** | Llama 3.1 70B, DeepSeek V3 MoE | Frontier performance, locally |
Every model runs privately, offline, with no subscription.
---
## Why local AI matters
When you use AI online, your words travel to a server in another country. Someone can read them. You pay per word. The service can shut down.
With Inference-X, your questions stay on your desk. The answer is computed by your own processor. Nothing leaves. Nothing is stored. It works without internet. It's free forever.
--- ---
## What makes it different ## What makes it different
Most inference engines add layers between the model and the hardware: frameworks, runtime allocators, intermediate buffers. Each layer degrades the model's signal. Most inference engines add layers between the model and the hardware: frameworks, runtime allocators, intermediate buffers, uniform precision pipelines. Each layer adds computational overhead that degrades the model's original signal.
Inference-X removes those layers. Inference-X removes those layers.
**Fused computation** — Dequantization and matrix multiply happen in a single instruction loop. No intermediate FP32 buffer. Output closer to the model's theoretical maximum. **Fused computation** — Dequantization and matrix multiply happen in a single instruction loop. No intermediate FP32 buffer. Fewer rounding operations means output closer to the model's theoretical FP32 maximum.
**Adaptive precision** — Each query is analyzed before inference. Simple questions get compressed early layers and full-precision decision layers. Complex reasoning gets full precision throughout. **Adaptive precision** — Each query is analyzed before inference. Simple questions get compressed early layers and full-precision decision layers. Complex reasoning gets full precision throughout. The model adapts its depth to the question — same file, same binary, different computational path.
**Surgical expert loading** — For Mixture-of-Experts models, only active experts exist in memory. A 1-trillion-parameter model runs on 64 GB of RAM. **Surgical expert loading** — For Mixture-of-Experts models, only active experts exist in memory. Inactive experts are evicted at the OS level. Result: a 1-trillion-parameter model runs on 17 GB of RAM. The signal path contains only what contributes to the current token.
The result: **the same model produces better output through a cleaner computation path.** A smaller model through Inference-X can match a larger model through a conventional engine. The result: **the same model produces higher-fidelity output through a cleaner computation path.** Or equivalently: a smaller model through Inference-X can match a larger model through a conventional engine.
→ [Full technical explanation](TECHNOLOGY.md) → [Full technical explanation](TECHNOLOGY.md)
--- ---
## How it works ## What it is
TCP/IP routes data packets to any network. Inference-X routes intelligence to any silicon. TCP/IP routes data packets to any network, any hardware, any destination. The protocol doesn't care about the wire.
One function call enters `kernel_dispatch.h`. On the other side: CPU, GPU, TPU, LPU, IPU, FPGA, DSP, or WSE. The model runs. The answer comes back. Inference-X routes intelligence to any silicon. The protocol doesn't care about the chip.
One function call enters `kernel_dispatch.h`. On the other side: CPU, GPU, TPU, LPU, IPU, FPGA, DSP, or WSE. The caller doesn't know. Doesn't need to. The model runs. The answer comes back.
``` ```
Model (any GGUF) → Inference-X (305 KB) → Silicon (any of 19 backends) → Response Model (any GGUF) → Inference-X (305 KB) → Silicon (any of 19 backends) → Response
``` ```
``` The model describes itself. The engine reads the description. The engine never assumes.
Architecture:
infer.cpp (570 lines) — Orchestrator. Chat templates. Server mode.
transformer_v6.h — Forward pass. Dense + MoE + MLA unified. ## Quick Start
kernel_dispatch.h — Routes GEMM to the right silicon.
moe_mla.h — Expert selection. Prefetch. Eviction. ```bash
gemm.h — Fused dequant × matmul kernels. git clone https://github.com/ElmadaniS/inference-x
backends.h — 19 hardware targets. One interface. cd inference-x
make
# Download a model (any GGUF from Hugging Face)
./inference-x model.gguf -p "Hello, world"
``` ```
12,571 lines of C++17. 6 architectures (Llama, Qwen2, Gemma2, Phi, DeepSeek MoE, MLA). 23 quantization formats. One binary. That's it. One binary. One command. Any model.
## Why it matters
Running a model today requires choosing a stack: CUDA for NVIDIA, ROCm for AMD, Metal for Apple, TensorRT for serving, vLLM for throughput, Ollama for local. Each stack locks you to a vendor, a way of thinking, and adds its own computational overhead between the model and the result.
Inference-X eliminates the stack. There is no stack. There's a model file, a binary, and your hardware — whatever it is.
```
GPU cluster: 1T parameters on 8× H100 ~5.6 kW, $200,000+/year
Inference-X: 1T parameters on 256 GB RAM ~300 W, €4,800/year
Same model. Cleaner output. 97% less cost.
```
This isn't about replacing GPUs. It's about making the choice of silicon irrelevant to the act of thinking — and getting *better* results from the silicon you already have.
## Who is this for
**Every organization that runs AI models — or wants to.**
| Sector | Problem | What IX does |
|--------|---------|-------------|
| **Healthcare** | Patient data can't leave the hospital. Cloud inference = compliance risk. | Air-gapped inference on hospital hardware. Zero network calls. HIPAA/GDPR by architecture. |
| **Defense & Government** | Sovereign AI requires sovereign infrastructure. | Runs on government-owned hardware. No vendor dependency. No telemetry. Auditable source. |
| **Finance** | Trading models need low latency and full auditability. | On-premise inference, deterministic output, no external calls. |
| **Telecom** | Edge inference at cell towers for real-time processing. | 305 KB binary deploys on edge hardware. Adaptive precision matches available power. |
| **Automotive** | In-vehicle AI needs minimal footprint and guaranteed response. | Runs on ARM/Snapdragon. No framework overhead. Fits in L2 cache. |
| **Startups** | GPU costs eat runway. $200K/year for inference infrastructure. | Same model quality at 97% lower cost. CPU-only. Scale when you're ready. |
| **Enterprise** | Vendor lock-in across NVIDIA, AMD, Intel, cloud providers. | 19 backends. One binary. Switch hardware without changing code. |
| **Research & Education** | Limited compute budgets. Students can't afford H100s. | Free under BSL-1.1. Run 14B models on a €20/month server. |
| **Embedded / IoT** | AI on microcontrollers with KB-level memory budgets. | Compiles for ESP32. Surgical loading keeps memory minimal. |
| **Cloud Providers** | Offering inference services at competitive margins. | Higher output quality per compute dollar. 19 backends = any customer hardware. |
Inference-X has zero friction with existing infrastructure. It doesn't replace your hardware — it makes your hardware work better.
## Get started
```bash
# Build (30 seconds)
git clone https://github.com/ElmadaniS/inference-x.git
cd inference-x && make -j$(nproc)
# Chat with any GGUF model
./inference-x model.gguf -i
# Or start a web interface
python3 web/ix_server.py
# Or run as an OpenAI-compatible API
./inference-x model.gguf --serve --port 8080
```
Three commands. No dependencies. No Docker. No Python packages. No GPU drivers. Just `make` and run.
---
## Benchmarks ## Benchmarks
AMD EPYC Rome · 17 GB RAM · 6 cores · CPU-only · €20/month server Real numbers on a €20/month AMD EPYC server. CPU-only. No GPU. Cold start.
| Model | Params | Quant | tok/s | Prefill | | Model | Params | Quant | tok/s |
|---|---|---|---|---| |-------|--------|-------|-------|
| SmolLM2 | 135M | Q8_0 | **130.23** | 87 ms | | SmolLM2 | 135M | Q8_0 | **130.23** |
| Qwen 2.5 | 3B | Q4_K_M | **3.85** | 16.5 s | | Llama 3.2 | 3B | Q4_K_M | **3.82** |
| Llama 3.2 | 3B | Q4_K_M | **3.82** | 3.8 s | | Qwen 2.5 | 3B | Q4_K_M | **3.85** |
| Mistral 7B | 7B | Q4_K_M | **2.06** | 39.2 s | | Mistral 7B | 7B | Q4_K_M | **2.06** |
| Llama 3.1 | 8B | Q4_K_M | **1.75** | 43.0 s | | Qwen 2.5 | 7B | Q4_K_M | **1.82** |
| DeepSeek R1 | 14B | Q4_K_M | **0.97** | 74.1 s | | Llama 3.1 | 8B | Q4_K_M | **1.75** |
| Gemma 2 | 9B | Q4_K_M | **1.28** |
| DS-R1 Qwen | 14B | Q4_K_M | **0.97** |
9 models · 4 architectures · Same binary · Zero configuration 9/10 architectures passing. Chat templates auto-detected. Zero manual configuration.
→ [Full benchmarks](BENCHMARKS.md) → [Full benchmark details](BENCHMARKS.md)
---
## Supported Hardware ## Supported Hardware
| Backend | Target | Status | | Backend | Silicon | Status |
|---|---|---| |---------|---------|--------|
| CPU AVX2/512 | Intel, AMD | ✅ Production | | CPU (AVX2/AVX-512) | Intel, AMD | ✅ Production |
| CUDA | NVIDIA GPU | ✅ Production | | CUDA | NVIDIA GPU | ✅ Production |
| ROCm | AMD GPU | ✅ Production | | ROCm | AMD GPU | ✅ Production |
| Metal | Apple Silicon | ✅ Production | | Metal | Apple Silicon | ✅ Production |
| Vulkan | Cross-platform | ✅ Production | | Vulkan | Cross-platform GPU | ✅ Production |
| ARM NEON | ARM (Pi, phones) | ✅ Production | | ARM NEON | ARM processors | ✅ Production |
| Snapdragon | Qualcomm | 🔶 Ready | | Snapdragon | Qualcomm (GPU+DSP+NEON) | 🔧 Ready |
| Hexagon HVX | Qualcomm DSP | 🔶 Ready | | Hexagon HVX | Qualcomm DSP | 🔧 Ready |
| TPU | Google | 🔶 Ready | | OpenCL | Cross-platform | 🔧 Ready |
| Inferentia | AWS | 🔶 Ready | | WebGPU | Browser | 🔧 Ready |
| Gaudi | Intel HPU | 🔶 Ready | | TPU | Google | 🔧 Ready |
| Maia | Microsoft | 🔶 Ready | | Inferentia | AWS | 🔧 Ready |
| SambaNova RDU | SambaNova | 🔶 Ready | | Gaudi | Intel HPU | 🔧 Ready |
| Graphcore IPU | Graphcore | 🔶 Ready | | Maia | Microsoft | 🔧 Ready |
| Groq LPU | Groq | 🔶 Ready | | SambaNova RDU | SambaNova | 🔧 Ready |
| Cerebras WSE | 850K cores | 🔶 Ready | | Graphcore IPU | Graphcore | 🔧 Ready |
| FPGA | Xilinx | 🔶 Ready | | Groq LPU | Groq | 🔧 Ready |
| WebGPU | Browser | 🔶 Ready | | FPGA (Xilinx) | Xilinx | 🔧 Ready |
| OpenCL | Universal | 🔶 Ready | | Cerebras WSE | Cerebras | 🔧 Ready |
The Makefile detects your hardware. You don't configure it.
--- ## Architecture
```
infer.cpp ← Entry point (571 lines)
├── runtime/
│ ├── gguf.h ← GGUF parser + config extraction
│ ├── tokenizer.h ← Tokenizer with byte-level BPE
│ ├── transformer_v6.h ← Universal forward pass
│ ├── attention.h ← GQA attention
│ ├── moe_mla.h ← MoE + MLA (DeepSeek V3)
│ ├── gemm.h ← Fused GEMV kernels
│ ├── kernels.h ← RMS norm, softmax, RoPE, SiLU
│ ├── kernel_dispatch.h ← Hardware routing layer
│ ├── server.h ← OpenAI-compatible API server
│ └── ...
├── core/
│ ├── iq_tables.h ← IQ quantization lookup tables
│ └── z_core.h ← Mathematical foundation
└── backends/
└── q4_kernels/ ← Per-hardware kernel implementations
```
One forward pass handles: dense transformers, Mixture-of-Experts, Multi-head Latent Attention, grouped-query attention, fused QKV tensors, and every combination.
→ [Detailed architecture](ARCHITECTURE.md) · [How the technology works](TECHNOLOGY.md)
## Features
- **Higher fidelity output** — Fused dequant+dot kernels eliminate intermediate buffers. Fewer rounding operations = output closer to the model's FP32 theoretical maximum.
- **Adaptive precision** — Shannon entropy analysis determines per-layer quantization. Simple queries run faster. Complex reasoning gets full depth. The model breathes.
- **Surgical expert loading** — MoE models load only active experts. 48× I/O reduction. Clean signal path with zero interference from unused parameters.
- **Universal model support** — LLAMA, QWEN2, PHI3, GEMMA2, DEEPSEEK, KIMI. Dense and MoE. The model changes, the protocol doesn't.
- **23 native quantization formats** — Q2_K through FP32. No format conversion. The engine speaks the model's native dialect.
- **19 hardware backends** — CPU, GPU, TPU, LPU, IPU, FPGA, DSP, WSE. One binary, every silicon.
- **305 KB binary** — Fits in L2 cache. The engine is invisible. You hear the model, not the framework.
- **Auto chat template** — ChatML, Llama 3, Mistral, Gemma, Phi-3, Kimi. Detected from GGUF metadata. Zero configuration.
- **OpenAI-compatible API** — `./inference-x model.gguf --serve` gives you `/v1/chat/completions`. Drop-in replacement.
- **Web interface** — Built-in chat UI. `python3 web/ix_server.py` and open your browser.
## API Server ## API Server
Start with `--serve 8080`. OpenAI-compatible API. Any client library works. ```bash
./inference-x model.gguf --serve --port 8080
```
Drop-in replacement for OpenAI:
```python ```python
from openai import OpenAI from openai import OpenAI
client = OpenAI(base_url="http://localhost:8080/v1", api_key="none") client = OpenAI(base_url="http://localhost:8080/v1", api_key="none")
resp = client.chat.completions.create( response = client.chat.completions.create(
model="local", model="local",
messages=[{"role": "user", "content": "Hello!"}], messages=[{"role": "user", "content": "Hello"}]
stream=True
) )
``` ```
Endpoints: `POST /v1/chat/completions` · `POST /v1/completions` · `GET /v1/models` · `GET /health`
---
## Features
- **Universal GGUF** — Any model, any architecture, auto-detected from metadata
- **Chat templates** — 7 formats auto-detected (Llama, ChatML, Alpaca, Gemma, Phi, Mistral, DeepSeek)
- **Multi-EOS** — Correct stop tokens for every architecture
- **Server mode** — OpenAI-compatible API, streaming, health check
- **Air-gapped** — No network calls during inference. No telemetry. Ever.
- **Zero configuration** — Download a model, run it. Templates, tokens, architecture: auto.
---
## Contributing ## Contributing
See [CONTRIBUTING.md](CONTRIBUTING.md). Run `make` to build. Run `make test` to test. Submit a PR. We welcome contributions:
We welcome contributions from everyone, regardless of experience level. If you're new to open source, look for issues tagged `good first issue`. - **Backends** — Port kernel implementations to new hardware
- **Models** — Add new architectures and quantization formats
- **Benchmarks** — Run benchmarks on diverse hardware
- **Documentation** — Tutorials, guides, translations
See [CONTRIBUTING.md](CONTRIBUTING.md) for details.
---
## License ## License
[BSL-1.1](LICENSE) — Business Source License [Business Source License 1.1](LICENSE) — Free for individuals, researchers, and small teams. Commercial use requires a license. Converts to open source in 2030.
**Free for**: individuals, researchers, students, open-source projects, organizations under $1M revenue. See [NOTICE](NOTICE) for full terms.
**Change date**: February 12, 2030 → [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0)
After 2030, everything becomes fully open source. Patents remain protected.
---
## Acknowledgments ## Acknowledgments
Built in Morocco for the world by [Salka Elmadani](https://x.com/ElmadaniSa13111). - **[Infomaniak](https://infomaniak.com)** — Swiss hosting partner
- **[Hetzner](https://hetzner.com)** — High-performance compute
> *The shortest path between model weights and output produces the cleanest signal. Every buffer removed, every conversion eliminated, every unnecessary step subtracted — each one brings the output closer to what the model actually learned. The path itself is the filter.* ---
**[Website](https://inference-x.com)** · **[Sponsor](https://git.inference-x.com/salka)** · **[Contact](mailto:Elmadani.SALKA@proton.me)** <p align="center">
<a href="https://inference-x.com">inference-x.com</a> ·
<a href="https://x.com/ElmadaniSa13111">@ElmadaniSa13111</a> ·
<a href="https://github.com/sponsors/ElmadaniS">Sponsor</a>
<br><br>
<em>Built in Morocco for the world.</em>
</p>

2
TECHNOLOGY.md
View File

@ -181,7 +181,7 @@ Kimi K2.5 on Inference-X:
## Try it ## Try it
```bash ```bash
git clone https://git.inference-x.com/salka/inference-x git clone https://github.com/ElmadaniS/inference-x
cd inference-x cd inference-x
make make
./inference-x model.gguf -p "Hello" ./inference-x model.gguf -p "Hello"

22
VISION.md
View File

@ -90,15 +90,19 @@ Intelligence doesn't need to be expensive. It needs to be *clean*.
--- ---
## Low-power inference ## Solar inference
Adaptive precision was built for signal quality. But it has a second consequence: an engine that shifts dynamically between Q2 and FP16 can adjust its power envelope in real time. Every hour, the Sun delivers more energy to Earth than humanity uses in a year. 173,000 terawatts, falling on deserts, rooftops, forgotten places.
Full precision when power is abundant. Compressed when it's constrained. Minimal when running on battery. If inference requires 5–15 kW per rack, you need solar farms and battery banks.
A standard inference rack draws 5–15 kW. Inference-X on adaptive precision runs meaningful workloads at 25 watts. That's the difference between needing a power plant and needing a panel. If inference requires 25 watts, you need a camping panel.
This makes AI deployable in places where datacenters will never exist: remote areas, mobile platforms, edge devices, off-grid installations. The engine adapts to whatever energy is available. Adaptive precision was built for a different reason. But it turns out: an engine that can dynamically shift between Q2 and FP16 is exactly what solar inference needs. When the Sun is high, full precision. At twilight, compressed. At night, minimal.
The engine breathes with the Sun like it breathes with the question.
The first solar deployment target is 2026. Anti-Atlas, Morocco. 320 days of sun per year. The nearest datacenter is 1,000 kilometers away.
--- ---
@ -109,8 +113,8 @@ We don't announce timelines. We announce results.
- The engine is done. 305 KB. Running in production. - The engine is done. 305 KB. Running in production.
- The technology page explains how it works: [TECHNOLOGY.md](TECHNOLOGY.md) - The technology page explains how it works: [TECHNOLOGY.md](TECHNOLOGY.md)
- The benchmarks are real: [BENCHMARKS.md](BENCHMARKS.md) - The benchmarks are real: [BENCHMARKS.md](BENCHMARKS.md)
- The documentation is live: [docs.inference-x.com](https://docs.inference-x.com) - The web interface is live: [inference-x.com](https://inference-x.com)
- The low-power adaptation is in development. - The solar adaptation is in development.
--- ---
@ -120,7 +124,7 @@ Every great infrastructure made something abundant that was once scarce. Aqueduc
The next abundance is intelligence. Not artificial. Not corporate. Not as-a-service. The next abundance is intelligence. Not artificial. Not corporate. Not as-a-service.
Just intelligence. Clean. Accessible. Powered by whatever energy is available — from a datacenter to a rooftop. Just intelligence. Clean. Accessible. Powered by whatever energy is available — from a datacenter to a star.
The model already knows. The engine just needs to get out of the way. The model already knows. The engine just needs to get out of the way.
@ -129,3 +133,5 @@ The model already knows. The engine just needs to get out of the way.
*Salka Elmadani* *Salka Elmadani*
*February 2026* *February 2026*
*Built in Morocco for the world.* *Built in Morocco for the world.*
◆

2
core/iq_tables.h
View File

@ -6,7 +6,7 @@
// //
// INTELLECTUAL PROPERTY PROTECTION: // INTELLECTUAL PROPERTY PROTECTION:
// - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026) // - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026)
// - GitHub: git.inference-x.com/salka/inference-x // - GitHub: github.com/ElmadaniS/inference-x
// - Author: Salka Elmadani | Morocco | Morocco // - Author: Salka Elmadani | Morocco | Morocco
// //
// MANUFACTURER NOTICE: Any manufacturer, company, or entity that // MANUFACTURER NOTICE: Any manufacturer, company, or entity that

2
core/iq_tables_ext.h
View File

@ -6,7 +6,7 @@
// //
// INTELLECTUAL PROPERTY PROTECTION: // INTELLECTUAL PROPERTY PROTECTION:
// - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026) // - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026)
// - GitHub: git.inference-x.com/salka/inference-x // - GitHub: github.com/ElmadaniS/inference-x
// - Author: Salka Elmadani | Morocco | Morocco // - Author: Salka Elmadani | Morocco | Morocco
// //
// MANUFACTURER NOTICE: Any manufacturer, company, or entity that // MANUFACTURER NOTICE: Any manufacturer, company, or entity that

2
core/z_core.h
View File

@ -6,7 +6,7 @@
// //
// INTELLECTUAL PROPERTY PROTECTION: // INTELLECTUAL PROPERTY PROTECTION:
// - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026) // - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026)
// - GitHub: git.inference-x.com/salka/inference-x // - GitHub: github.com/ElmadaniS/inference-x
// - Author: Salka Elmadani | Morocco | Morocco // - Author: Salka Elmadani | Morocco | Morocco
// //
// MANUFACTURER NOTICE: Any manufacturer, company, or entity that // MANUFACTURER NOTICE: Any manufacturer, company, or entity that

BIN
ifrane.pdf Normal file
View File

Binary file not shown.

4
infer.cpp
View File

@ -6,7 +6,7 @@
// //
// INTELLECTUAL PROPERTY PROTECTION: // INTELLECTUAL PROPERTY PROTECTION:
// - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026) // - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026)
// - GitHub: git.inference-x.com/salka/inference-x // - GitHub: github.com/ElmadaniS/inference-x
// - Author: Salka Elmadani | Morocco | Morocco // - Author: Salka Elmadani | Morocco | Morocco
// //
// MANUFACTURER NOTICE: Any manufacturer, company, or entity that // MANUFACTURER NOTICE: Any manufacturer, company, or entity that
@ -39,7 +39,7 @@ static void ix_print_banner() {
fprintf(stderr, " ║ Inference-X — Universal Inference Protocol ║\n"); fprintf(stderr, " ║ Inference-X — Universal Inference Protocol ║\n");
fprintf(stderr, " ║ Copyright (C) 2025-2026 Salka Elmadani ║\n"); fprintf(stderr, " ║ Copyright (C) 2025-2026 Salka Elmadani ║\n");
fprintf(stderr, " ║ Licensed under BSL-1.1 | Morocco ║\n"); fprintf(stderr, " ║ Licensed under BSL-1.1 | Morocco ║\n");
fprintf(stderr, " ║ https://inference-x.com | git.inference-x.com/salka/inference-x║\n"); fprintf(stderr, " ║ https://inference-x.com | github.com/ElmadaniS/inference-x║\n");
fprintf(stderr, " ╚═══════════════════════════════════════════════════════════╝\n"); fprintf(stderr, " ╚═══════════════════════════════════════════════════════════╝\n");
fprintf(stderr, "\n"); fprintf(stderr, "\n");
} }

2
runtime/attention.h
View File

@ -6,7 +6,7 @@
// //
// INTELLECTUAL PROPERTY PROTECTION: // INTELLECTUAL PROPERTY PROTECTION:
// - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026) // - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026)
// - GitHub: git.inference-x.com/salka/inference-x // - GitHub: github.com/ElmadaniS/inference-x
// - Author: Salka Elmadani | Morocco | Morocco // - Author: Salka Elmadani | Morocco | Morocco
// //
// MANUFACTURER NOTICE: Any manufacturer, company, or entity that // MANUFACTURER NOTICE: Any manufacturer, company, or entity that

2
runtime/backends.h
View File

@ -6,7 +6,7 @@
// //
// INTELLECTUAL PROPERTY PROTECTION: // INTELLECTUAL PROPERTY PROTECTION:
// - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026) // - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026)
// - GitHub: git.inference-x.com/salka/inference-x // - GitHub: github.com/ElmadaniS/inference-x
// - Author: Salka Elmadani | Morocco | Morocco // - Author: Salka Elmadani | Morocco | Morocco
// //
// MANUFACTURER NOTICE: Any manufacturer, company, or entity that // MANUFACTURER NOTICE: Any manufacturer, company, or entity that

2
runtime/expert_mmap.h
View File

@ -6,7 +6,7 @@
// //
// INTELLECTUAL PROPERTY PROTECTION: // INTELLECTUAL PROPERTY PROTECTION:
// - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026) // - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026)
// - GitHub: git.inference-x.com/salka/inference-x // - GitHub: github.com/ElmadaniS/inference-x
// - Author: Salka Elmadani | Morocco | Morocco // - Author: Salka Elmadani | Morocco | Morocco
// //
// MANUFACTURER NOTICE: Any manufacturer, company, or entity that // MANUFACTURER NOTICE: Any manufacturer, company, or entity that

2
runtime/gemm.h
View File

@ -6,7 +6,7 @@
// //
// INTELLECTUAL PROPERTY PROTECTION: // INTELLECTUAL PROPERTY PROTECTION:
// - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026) // - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026)
// - GitHub: git.inference-x.com/salka/inference-x // - GitHub: github.com/ElmadaniS/inference-x
// - Author: Salka Elmadani | Morocco | Morocco // - Author: Salka Elmadani | Morocco | Morocco
// //
// MANUFACTURER NOTICE: Any manufacturer, company, or entity that // MANUFACTURER NOTICE: Any manufacturer, company, or entity that

2
runtime/gguf.h
View File

@ -6,7 +6,7 @@
// //
// INTELLECTUAL PROPERTY PROTECTION: // INTELLECTUAL PROPERTY PROTECTION:
// - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026) // - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026)
// - GitHub: git.inference-x.com/salka/inference-x // - GitHub: github.com/ElmadaniS/inference-x
// - Author: Salka Elmadani | Morocco | Morocco // - Author: Salka Elmadani | Morocco | Morocco
// //
// MANUFACTURER NOTICE: Any manufacturer, company, or entity that // MANUFACTURER NOTICE: Any manufacturer, company, or entity that

2
runtime/kernel_dispatch.h
View File

@ -6,7 +6,7 @@
// //
// INTELLECTUAL PROPERTY PROTECTION: // INTELLECTUAL PROPERTY PROTECTION:
// - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026) // - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026)
// - GitHub: git.inference-x.com/salka/inference-x // - GitHub: github.com/ElmadaniS/inference-x
// - Author: Salka Elmadani | Morocco | Morocco // - Author: Salka Elmadani | Morocco | Morocco
// //
// MANUFACTURER NOTICE: Any manufacturer, company, or entity that // MANUFACTURER NOTICE: Any manufacturer, company, or entity that

2
runtime/kernels.h
View File

@ -6,7 +6,7 @@
// //
// INTELLECTUAL PROPERTY PROTECTION: // INTELLECTUAL PROPERTY PROTECTION:
// - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026) // - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026)
// - GitHub: git.inference-x.com/salka/inference-x // - GitHub: github.com/ElmadaniS/inference-x
// - Author: Salka Elmadani | Morocco | Morocco // - Author: Salka Elmadani | Morocco | Morocco
// //
// MANUFACTURER NOTICE: Any manufacturer, company, or entity that // MANUFACTURER NOTICE: Any manufacturer, company, or entity that

2
runtime/moe_mla.h
View File

@ -6,7 +6,7 @@
// //
// INTELLECTUAL PROPERTY PROTECTION: // INTELLECTUAL PROPERTY PROTECTION:
// - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026) // - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026)
// - GitHub: git.inference-x.com/salka/inference-x // - GitHub: github.com/ElmadaniS/inference-x
// - Author: Salka Elmadani | Morocco | Morocco // - Author: Salka Elmadani | Morocco | Morocco
// //
// MANUFACTURER NOTICE: Any manufacturer, company, or entity that // MANUFACTURER NOTICE: Any manufacturer, company, or entity that

2
runtime/tokenizer.h
View File

@ -6,7 +6,7 @@
// //
// INTELLECTUAL PROPERTY PROTECTION: // INTELLECTUAL PROPERTY PROTECTION:
// - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026) // - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026)
// - GitHub: git.inference-x.com/salka/inference-x // - GitHub: github.com/ElmadaniS/inference-x
// - Author: Salka Elmadani | Morocco | Morocco // - Author: Salka Elmadani | Morocco | Morocco
// //
// MANUFACTURER NOTICE: Any manufacturer, company, or entity that // MANUFACTURER NOTICE: Any manufacturer, company, or entity that

2
runtime/transformer_v6.h
View File

@ -6,7 +6,7 @@
// //
// INTELLECTUAL PROPERTY PROTECTION: // INTELLECTUAL PROPERTY PROTECTION:
// - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026) // - INPI eSoleau deposit: 7phf-Ueye-2nWr-Vsgu (16/02/2026)
// - GitHub: git.inference-x.com/salka/inference-x // - GitHub: github.com/ElmadaniS/inference-x
// - Author: Salka Elmadani | Morocco | Morocco // - Author: Salka Elmadani | Morocco | Morocco
// //
// MANUFACTURER NOTICE: Any manufacturer, company, or entity that // MANUFACTURER NOTICE: Any manufacturer, company, or entity that

4
web/ix_server.py
View File

@ -1,7 +1,7 @@
#!/usr/bin/env python3 #!/usr/bin/env python3
""" """
IX Web — Web interface for Inference-X IX Web — Web interface for Inference-X
https://git.inference-x.com/salka/inference-x https://github.com/ElmadaniS/inference-x
Zero dependencies. Pure Python stdlib. Zero dependencies. Pure Python stdlib.
Serves the IX Web chat UI and wraps the IX binary with an OpenAI-compatible API. Serves the IX Web chat UI and wraps the IX binary with an OpenAI-compatible API.
@ -413,7 +413,7 @@ class IXHandler(http.server.BaseHTTPRequestHandler):
def main(): def main():
parser = argparse.ArgumentParser( parser = argparse.ArgumentParser(
description="IX Web — Web interface for Inference-X", description="IX Web — Web interface for Inference-X",
epilog="https://git.inference-x.com/salka/inference-x", epilog="https://github.com/ElmadaniS/inference-x",
) )
parser.add_argument("--port", type=int, default=DEFAULT_PORT, help=f"Port (default: {DEFAULT_PORT})") parser.add_argument("--port", type=int, default=DEFAULT_PORT, help=f"Port (default: {DEFAULT_PORT})")
parser.add_argument("--host", default="0.0.0.0", help="Bind address (default: 0.0.0.0)") parser.add_argument("--host", default="0.0.0.0", help="Bind address (default: 0.0.0.0)")