release: v2.1.0 — promote from rc1 to stable

Promote v2.1.0rc1 to a full stable release. All production-hardening
fixes from the rc1 cycle are included: honest is_quantized reporting,
architecture fallback fixes, pre-quantized repo detection, expanded
model family support, real CI pipeline, and PEP 621 packaging.
Co-authored-by: monkeycode-ai <monkeycode-ai@chaitin.com>

Author: MonkeyCode-AI

CHANGELOG.md

@@ -4,7 +4,7 @@ All notable changes to QuantLLM are recorded here. The format follows
 [Keep a Changelog](https://keepachangelog.com/en/1.1.0/) and the project
 adheres to [Semantic Versioning](https://semver.org/).
 
-## [Unreleased] — production hardening on top of v2.1.0rc1
+## [2.1.0] — 2026-07-29
 
 ### Fixed
 
@@ -93,3 +93,6 @@ Initial public release of the `turbo()` API and the GGUF / ONNX / MLX
 export pipeline. See the GitHub
 [releases page](https://github.com/codewithdark-git/QuantLLM/releases/tag/v2.0.0)
 for the full notes.
+
+[2.1.0]: https://github.com/codewithdark-git/QuantLLM/compare/v2.0.0...v2.1.0
+[2.0.0]: https://github.com/codewithdark-git/QuantLLM/releases/tag/v2.0.0

README.md

@@ -1,7 +1,7 @@
 <div align="center">
   <img src="docs/images/1.png" alt="QuantLLM Logo" />
 
-  # 🚀 QuantLLM v2.1 (pre-release)
+  # 🚀 QuantLLM v2.1
 
   **The Ultra-Fast LLM Quantization & Export Library**
 
@@ -139,7 +139,7 @@ Llama 2/3, Mistral, Mixtral, Qwen 1/2, Phi 1/2/3, Gemma, Falcon, DeepSeek, Yi, S
 
 ```
 ╔════════════════════════════════════════════════════════════╗
-║   🚀 QuantLLM v2.1.0rc1                                    ║
+║   🚀 QuantLLM v2.1.0                                        ║
 ║   Ultra-fast LLM Quantization & Export                     ║
 ║   ✓ GGUF  ✓ ONNX  ✓ MLX  ✓ SafeTensors                     ║
 ╚════════════════════════════════════════════════════════════╝

docs/api/turbo.md

@@ -126,7 +126,7 @@ When `verbose=True` (default), you'll see:
 
 ```
 ╔════════════════════════════════════════════════════════════╗
-║  🚀 QuantLLM v2.1.0rc1                                        ║
+║  🚀 QuantLLM v2.1.0                                            ║
 ╚════════════════════════════════════════════════════════════╝
 
 📊 Loading: meta-llama/Llama-3.2-3B

docs/conf.py

@@ -3,7 +3,7 @@
 project = "QuantLLM"
 copyright = "2024, Dark Coder"
 author = "Dark Coder"
-release = "2.1.0rc1"
+release = "2.1.0"
 
 # Extensions
 extensions = [

docs/installation.md

@@ -91,7 +91,7 @@ Expected output:
 ```
 ╔════════════════════════════════════════════════════════════╗
 ║                                                            ║
-║   🚀 QuantLLM v2.1.0rc1                                       ║
+║   🚀 QuantLLM v2.1.0                                           ║
 ║   Ultra-fast LLM Quantization & Export                     ║
 ║                                                            ║
 ║   ✓ GGUF  ✓ ONNX  ✓ MLX  ✓ SafeTensors                     ║

docs/quickstart.md

@@ -192,7 +192,7 @@ quantllm.show_banner()
 ```
 ╔════════════════════════════════════════════════════════════╗
 ║                                                            ║
-║   🚀 QuantLLM v2.1.0rc1                                       ║
+║   🚀 QuantLLM v2.1.0                                           ║
 ║   Ultra-fast LLM Quantization & Export                     ║
 ║                                                            ║
 ║   ✓ GGUF  ✓ ONNX  ✓ MLX  ✓ SafeTensors                     ║

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "quantllm"
-version = "2.1.0rc1"
+version = "2.1.0"
 description = "Ultra-fast LLM quantization, fine-tuning, and deployment with one line of code"
 readme = "README.md"
 license = {text = "MIT"}

quantllm/__init__.py

@@ -77,7 +77,7 @@
 # Configure logging (minimal by default)
 configure_logging("WARNING")
 
-__version__ = "2.1.0rc1"
+__version__ = "2.1.0"
 __title__ = "QuantLLM"
 __description__ = "Ultra-fast LLM Quantization & Export (GGUF, ONNX, MLX)"
 __author__ = "Dark Coder"