Search Results for "decoding"
Sort By:
210 projects for "decoding" with 1 filter applied:
-
Train ML Models With SQL You Already KnowBuild and deploy ML models using familiar SQL. Automate data prep with built-in Gemini. Query 1 TB and store 10 GB free monthly.
-
Custom VMs From 1 to 96 vCPUs With 99.95% UptimeLive migration and automatic failover keep workloads online through maintenance. One free e2-micro VM every month.
-
1
h265web.js
A HEVC/H.265 Web Player
...The project includes components for parsing H.265 bitstreams into NAL units and decoding them into raw YUV frames that can then be rendered onto a canvas or processed further. Its architecture separates parsing, decoding, and rendering, giving developers fine-grained control over how video data is handled and displayed. The system is designed to work with streaming data, allowing incremental feeding of video chunks and real-time decoding, which is useful for surveillance, streaming platforms, and custom media pipelines. -
2
DeepSpec
A full-stack codebase for training and evaluating speculative decoding
...The evaluation pipeline measures speculative decoding performance across benchmark tasks such as math, coding, instruction-following, and chat-style datasets. Overall, it is useful for researchers and engineers studying faster language model inference through speculative decoding methods. -
3
DFlash
Block Diffusion for Ultra-Fast Speculative Decoding
DFlash is an open-source framework for ultra-fast speculative decoding using a lightweight block diffusion model to draft text in parallel with a target large language model, dramatically improving inference speed without sacrificing generation quality. It acts as a “drafter” that proposes likely continuations which the main model then verifies, enabling significant throughput gains compared to traditional autoregressive decoding methods that generate token by token. -
4
io-ts
Runtime type system for IO decoding/encoding
Runtime type system for IO decoding/encoding -
Build Agents and Models on One PlatformGemini Enterprise Agent Platform is Google Cloud's comprehensive platform for developers to build, scale, govern, and optimize agents and models. Choose from Google's most advanced models and third-party models like Anthropic's Claude Model Family.
-
5
R-KV
Redundancy-aware KV Cache Compression for Reasoning Models
R-KV is an open-source research project that focuses on improving the efficiency of large language model inference through key-value cache compression techniques. Modern transformer models rely heavily on KV caches during autoregressive decoding, which store intermediate attention states to accelerate generation. However, these caches can consume large amounts of memory, especially in reasoning-oriented models with long context windows. R-KV introduces a method for compressing the KV cache during decoding, allowing models to maintain reasoning performance while reducing memory consumption and computational overhead. ... -
6
FlashMLA
FlashMLA: Efficient Multi-head Latent Attention Kernels
FlashMLA is a high-performance decoding kernel library designed especially for Multi-Head Latent Attention (MLA) workloads, targeting NVIDIA Hopper GPU architectures. It provides optimized kernels for MLA decoding, including support for variable-length sequences, helping reduce latency and increase throughput in model inference systems using that attention style. The library supports both BF16 and FP16 data types, and includes a paged KV cache implementation with a block size of 64 to efficiently manage memory during decoding. -
7
he
A robust HTML entity encoder/decoder written in JavaScript
...It emphasizes correctness, security (avoiding common pitfalls when decoding entities), and broad compatibility (browser, Node.js). The API is straightforward, documented, and used across modules that require reliable entity handling (e.g., for HTML sanitization or templating). Its longevity and adoption make it a go-to library for developers needing entity handling beyond minimal ad-hoc hacks. -
8
io-ts
Runtime type system for IO decoding/encoding
io-ts is a TypeScript runtime type system for decoding, validating, and encoding data at application boundaries. It solves a common TypeScript limitation: static types disappear at runtime, but external inputs still need to be checked safely. The library lets developers define codecs that represent both runtime validation logic and static TypeScript types. This approach is especially useful for APIs, configuration files, JSON payloads, database results, and any data that enters a program from an unknown source. io-ts integrates closely with functional programming patterns and the fp-ts ecosystem. ... -
9
Jessibuca
Jessibuca is an open source pure H5 live streaming player
Jessibuca is an open-source HTML5 live streaming player that runs entirely in the browser using WebAssembly to decode audio and video streams without requiring plugins. It compiles native decoding libraries into JavaScript, enabling high-performance playback across desktop and mobile devices. The player supports modern streaming protocols and codecs, allowing it to handle real-time video feeds with low latency and stable rendering. Its architecture is modular, separating decoding, demuxing, and rendering to optimize performance and scalability. ... -
99.99% Uptime for MySQL and PostgreSQL DatabasesCloud SQL Enterprise Plus delivers near-zero downtime with 35 days of point-in-time recovery. Supports MySQL, PostgreSQL, and SQL Server.
-
10
PyAV
Pythonic bindings for FFmpeg's libraries
...The library is designed for precise media manipulation, making it suitable for applications that require frame-level access or custom processing pipelines. PyAV integrates well with other Python libraries such as NumPy and Pillow, allowing seamless data transformation and analysis. It supports decoding, encoding, filtering, and resampling operations, covering a wide range of multimedia tasks. While powerful, it requires a solid understanding of FFmpeg concepts, as it prioritizes flexibility and control over abstraction. Overall, PyAV is a robust tool for developers building advanced video and audio processing systems in Python. -
11
MiniCPM4.1
Achieving 3+ generation speedup on reasoning tasks
MiniCPM4.1 is an enhanced iteration of the MiniCPM4 architecture, introducing improvements in reasoning capabilities, inference speed, and hybrid operation modes that allow dynamic switching between deep reasoning and standard generation. It builds upon the same efficiency-focused philosophy but further optimizes decoding performance, achieving substantial speed gains in reasoning-intensive tasks while maintaining high-quality outputs. One of its key innovations is the hybrid reasoning mode, which allows developers to control whether the model engages in deeper reasoning processes or faster responses depending on the use case. The model also supports both dense and sparse attention mechanisms, enabling more efficient computation depending on the selected inference framework. ... -
12
Cheshire
Clojure JSON and JSON SMILE (binary json format) encoding/decoding
Cheshire is a high-performance JSON encoding and decoding library for Clojure, built atop Jackson to combine fast parsing/generation with advanced support for types like Dates, UUIDs, Sets, Symbols, and the binary JSON SMILE format. Custom encoding is supported from 2.0.0 and up if you encounter a bug, please open a GitHub issue. From 5.0.0 onwards, custom encoding has been moved to be part of the core namespace (not requiring a namespace change). -
13
iced
Blazing fast and correct x86/x64 disassembler, assembler, decoder, etc
iced is a powerful and feature-rich disassembly and assembly library for x86 and x64 architectures, designed to provide accurate decoding, encoding, and formatting of machine instructions. It supports multiple programming languages, including C#, Rust, and Python, making it accessible to a wide range of developers. The library offers both disassembly and assembly capabilities, allowing users to convert between machine code and human-readable instructions in both directions. -
14
Decimen Optical Transfer
Fountain-coded QR file transfer
...Fountain coding makes the stream tolerant of missed, blurred, duplicated, and out-of-order frames without requesting retransmission. Each QR frame includes session and file metadata, allowing a receiver to join an active stream or reset when a new session begins. QR decoding runs through zxing-cpp compiled to WebAssembly and uses workers to process camera frames efficiently. Configurable settings control payload size, transmission speed, QR density, capture resolution, and worker count, while a final hash verifies the recovered file. -
15
Moquette Project
Java MQTT lightweight broker
...Simply embeddable in your IoT projects. Moquette aims to be a MQTT compliant broker. The broker supports QoS 0, QoS 1 and QoS 2. Its designed to be evented, uses Netty for the protocol encoding and decoding part. -
16
Tiktoken
tiktoken is a fast BPE tokeniser for use with OpenAI's models
...It also offers extension mechanisms so that custom encodings can be registered. Internally, it includes the core tokenizer logic (often implemented in Rust or efficient lower-level code), APIs for encoding, decoding, and counting tokens, and binding layers to Python (and sometimes other languages) for easy use. -
17
MiniCPM-V
See, hear, and interact on your local devices
...The larger model uses Lookahead Expert Scheduling to accelerate inference on memory-constrained devices. Quantized versions reduce hardware requirements, while speculative decoding further improves generation speed. The broader MiniCPM family also includes MiniCPM-o models for real-time audio, video, and speech interactions. Developers can use pretrained checkpoints, inference frameworks, and fine-tuning resources to build multimodal applications. -
18
PcapPlusPlus
Multiplatform C++ library for parsing and crafting of network packets
...PcapPlusPlus enables capturing and sending network packets through easy-to-use C++ wrappers for the most popular packet processing engines such as libpcap, WinPcap, NPcap, DPDK and PF_RING. PcapPlusPlus enables decoding and forging capabilities for a large variety of network protocols. PcapPlusPlus is designed to be efficient and lightweight. It enables amazingly fast packet processing with minimum overhead. PcapPlusPlus is fully supported on Windows, MacOS, Linux, Android and FreeBSD. You can download pre-built binaries for each platform or build it from source. ... -
19
NanoJev
A nano replica of Jev: parallel decisions, dynamic candidates
NanoJev is an open-source 0.6B-parameter decision model created as a compact reproduction of Jev-style parallel decision making. It takes states and questions as input and returns complete probability distributions without decoding output tokens one by one. The project supports dynamic candidate sets so the available actions can change from one decision to the next. One shared checkpoint has been tested across ViZDoom aiming tasks, maze navigation, and Snake. The repository includes datasets, configurations, research results, demonstrations, and an end-to-end training pipeline. ... -
20
Lucebox
Fast LLM speculative inference server for consumer hardware
Lucebox is a local LLM inference server built for fast generation on consumer hardware. It focuses on custom kernels, speculative prefill, speculative decoding, and model-specific optimizations rather than a generic one-size-fits-all runtime. The project includes a native C++ HTTP server with an OpenAI-compatible API, making it usable with tools that already speak the Chat Completions format. It supports CUDA and ROCm workflows, with Docker images for NVIDIA and AMD GPU setups. The repository also includes harnesses for testing compatibility with clients such as Claude Code, Codex, OpenCode, Hermes, Pi, OpenClaw, and Open WebUI. ... -
21
SwiftFFmpeg
A Swift wrapper for the FFmpeg API
SwiftFFmpeg is a Swift wrapper for FFmpeg that provides direct access to multimedia processing capabilities within Apple development environments. It exposes FFmpeg’s libraries through Swift-friendly APIs, enabling developers to perform encoding, decoding, and stream manipulation tasks. The project is designed to maintain performance while improving usability through modern language features. It supports working with media containers, codecs, and frames at a low level, offering fine control over processing pipelines. SwiftFFmpeg integrates seamlessly with iOS, macOS, and other Apple platforms, making it suitable for native app development. ... -
22
Requests
A simple, yet elegant, HTTP library.
Requests is the de facto HTTP library for Python—simple, elegant, and human-friendly. It wraps urllib3 to provide intuitive methods for sending HTTP/1.1 requests, handling sessions, cookies, redirects, authentication, proxies, and more. -
23
QOI
The “Quite OK Image Format” for fast, lossless image compression
QOI, short for “Quite OK Image,” is a minimalist lossless image compression format designed to achieve extremely fast encoding and decoding speeds while maintaining compression ratios comparable to PNG. The project was created as a simpler alternative to complex legacy image formats, using an intentionally lightweight specification that can be implemented in only a few hundred lines of code. QOI focuses on predictable O(n) performance and efficient pixel encoding strategies without relying on heavyweight compression algorithms like DEFLATE. ... -
24
WebP Codec
Library to encode and decode images in WebP format
libwebp is the reference codec library for Google’s WebP image format, providing both encoding and decoding along with command-line tools. It supplies cwebp to compress images into WebP and dwebp to decompress them back, making it easy to test quality/size trade-offs across presets and tuning parameters. The GitHub repository is a mirror; the canonical source of truth lives on Chromium’s git, and developer docs are hosted on WebP’s portal. -
25
mllm
Fast Multimodal LLM on Mobile Devices
...Implemented primarily in C and C++, it is designed to operate with minimal external dependencies while taking advantage of hardware-specific acceleration technologies such as ARM NEON and x86 AVX2 instructions. The system supports multiple optimization techniques including quantization, pruning, and speculative decoding to improve performance while reducing computational overhead. It also provides tools to convert models from popular formats like PyTorch checkpoints into optimized runtime formats that can be executed on supported hardware platforms.
