Filter by tags:
#Testing
#Software Engineering (SW)
#Database
#Security
#Bio
#MOE
#Robotics
#OCR
#TTS
#MCP
#Multimodal
#Machine Translation
#RL
#CV
#Audio
#ML
#Coding
#Image Generation
#AGI
#Agentic AI
#Infrastructure
#AI Agents
#LLMs
2026Fly.io

Sprites

Paid
Cloud Computing
AI Infrastructure

Sprites are persistent, hardware-isolated Linux environments designed for running arbitrary code with stateful checkpoint and restore capabilities. They function as dedicated microVMs that hibernate when idle and wake automatically, making them ideal for AI agents, development environments, and secure code execution.

Persistent Linux environments with full ext4 filesystem support
Hardware-level isolation using Firecracker microVMs
Transactional checkpoint and restore functionality
RegionUnited States
2026Microsoft

winapp CLI v0.5.0

Open Source
windows-development
cli

The Windows App Development CLI (winapp CLI) is a unified command-line interface designed to streamline the development of Windows applications. It simplifies complex tasks such as managing Windows SDKs, handling MSIX packaging, generating app identity, and configuring manifests and certificates for various app frameworks.

Unified command-line interface for SDK management and app packaging
Simplified package identity workflow for debugging without full MSIX packaging
Automated generation of manifests, assets, and development certificates
PricingFree
Version0.5.0
2026Microsoft

dynwinrt v0.1.0-preview.18

Open Source
Featured
WinRT
Node.js

dynwinrt is a dynamic Windows Runtime (WinRT) API projection for Node.js and Electron that allows developers to call native Windows APIs directly from JavaScript or TypeScript. It eliminates the need for writing native C++ addons or complex build steps by reading .winmd metadata and invoking COM vtables at runtime via libffi.

Dynamic invocation of WinRT APIs via libffi
Automatic generation of typed JavaScript and TypeScript wrappers
No requirement for native C++ addons or node-gyp
PricingFree
Version0.1.0-preview.18
2026DeepSeek AI

DeepSeek-V4-Flash-0731 v0731

Open Source
LLM
Agentic AI

DeepSeek-V4-Flash-0731 is an official, re-post-trained version of the DeepSeek-V4-Flash model, featuring significantly enhanced agentic capabilities and a built-in DSpark speculative decoding module. It is designed for high-efficiency, large-scale tasks, offering competitive performance in coding, reasoning, and multi-step tool use.

1 million token context window
384K maximum output token length
DSpark speculative decoding module for enhanced throughput
Version0731
RegionChina
2026Together AI

ThunderAgent

Open Source
AI Infrastructure
Inference Optimization

ThunderAgent is a program-aware scheduling system designed to optimize high-throughput agentic inference. By treating agentic workflows as schedulable programs rather than independent requests, it effectively eliminates KV cache thrashing and improves multi-node scaling.

Program-level scheduling to eliminate KV cache thrashing
Global waiting queue for efficient multi-node load balancing
Drop-in compatibility with OpenAI-compatible inference backends
PricingFree
RegionUnited States
2026Datalab

Marker v2.0.0

Open Source
Featured
OCR
Document Intelligence

Marker is a high-performance document intelligence pipeline designed to convert PDFs, images, and various office documents into clean Markdown, JSON, HTML, or chunks. It leverages advanced deep learning models, including the Surya VLM, to handle complex layouts, tables, equations, and multi-language text extraction with high accuracy.

Converts PDF, image, PPTX, DOCX, XLSX, HTML, and EPUB files
Formats tables, equations, inline math, and code blocks
Extracts and saves images while removing headers/footers
Version2.0.0
RegionUnited States
2026Vercel

Scriptc v0.0.17

Open Source
TypeScript
Compiler

scriptc is a TypeScript-to-native compiler that transforms ordinary TypeScript code into small, fast, self-contained executables. It eliminates the need for Node.js, V8, or any embedded JavaScript engine for statically compiled code, resulting in significantly faster startup times and smaller binary sizes.

Zero-runtime native compilation for static TypeScript
Extremely fast startup (~2.4ms) and small binary footprint (~170-200KB)
Three-tier execution model: Static (native), Dynamic (QuickJS-ng), and Rejected
PricingFree
Version0.0.17
2026Tencent

AngelSpec v0.1.0

Open Source
Speculative Decoding
LLM Inference

AngelSpec is a unified, torch-native training framework designed for speculative decoding, supporting both autoregressive Multi-Token Prediction (MTP) and block-parallel drafting architectures. It enables independent scaling of inference and training through a disaggregated architecture, facilitating high-performance model deployment.

Unified training pipeline for 6 draft architectures (DFly, DFlash, DFlare, Eagle3, DSpark, MTP)
Disaggregated training architecture with independent scaling for inference and optimization
Support for long-context training up to 128k tokens via Ulysses sequence parallelism
PricingFree
Version0.1.0
2026Empero

Qwythos-27B-v1 v1

Open Source
reasoning
multimodal

Qwythos-27B-v1 is an open-weight, full-parameter reasoning model built on the Qwen3.5-27B base. It is designed for advanced technical tasks, featuring native multi-token prediction, a full vision tower, and a 1M-token context window.

1,048,576-token (1M) context window via YaRN
Native multi-token prediction (MTP) head preserved
Full vision tower integration for multimodal tasks
PricingFree (Open Source)
Version1
2026kvcache-ai & Moonshot AI

AgentENV v0.1.0

Open Source
AI Agents
Virtualization

AgentENV (AENV) is a distributed, open-source platform designed for running isolated AI agent environments at scale. It utilizes Firecracker microVMs to provide secure, kernel-level isolation while enabling rapid snapshotting, pausing, and forking of environments to support efficient agentic reinforcement learning workflows.

Firecracker microVM-based isolation for secure, kernel-level sandboxing
Instant environment state management with sub-100ms snapshot, pause, and resume capabilities
Native support for environment forking to enable parallel agent workflows
PricingFree & Open Source
Version0.1.0
2026OpenAI

Codex Security v0.1.1

Paid
Featured
Security
DevSecOps

Codex Security is an application security agent designed to help development teams identify, validate, and remediate vulnerabilities within their codebases. It provides both a CLI and a TypeScript SDK to scan repositories, review changes, and propose actionable fixes for security issues.

Automated vulnerability scanning for repositories and scoped folders
AI-powered validation of findings to reduce false positives
Actionable patch suggestions for identified security issues
Version0.1.1
RegionUnited States
2026Liquid AI

LFM2.5-Encoder-350M v2.5

Open Source
encoder
bidirectional

LFM2.5-Encoder-350M is a multilingual bidirectional encoder model built on the LFM2 hybrid architecture. It is designed for high-performance, on-device tasks such as text classification, retrieval, and semantic similarity across 15 languages.

8,192-token context window
Multilingual support across 15 languages
Efficient LFM2 hybrid backbone with gated convolutions
PricingFree to download, run, and fine-tune
Version2.5

🚀 Join the AI dev community — follow us everywhere

© 2026 MARKTECHPOST AI MEDIA INC. All rights reserved.Terms & ConditionsPrivacy Policy
Beta Mode