#MoE architecture
35 related articles
Tech Frontiers·2 min
LFM2.5-8B-A1B: A MoE Model with 1.5B Active Parameters Delivering 4x Its Weight Class Performance
Liquid AI releases LFM2.5-8B-A1B, a MoE model with 8B total params but only 1.5B active, matching 6B-class models in tool calling. Supports 128K context, local deployment, multilingual, with SGLang Day-0 support.
Read more →
Industry Insights·2 min
AMD MI355X Beats B200: Full-Stack Optimization Breakdown for 5% Lower TCO on DeepSeek-R1 Inference
AMD Instinct MI355X achieves 5% lower TCO than NVIDIA B200 on DeepSeek-R1 disaggregated inference via SGLang+MoRI full-stack optimization with 1.25x per-GPU throughput.
Read more →
Tech Frontiers·1 min
Cloudflare Contributes Critical KV Cache and Mooncake Fixes to SGLang
Cloudflare contributes decode KV cache offload and Mooncake recovery fixes to SGLang, resolving garbled output under high concurrency for Kimi K2.6 and enabling automatic fault recovery in distributed inference.
Read more →
Industry Insights
Deep Dive into Three Major LLM Career …
·3 min
Deep Dive into Three Major LLM Career Paths: Requirements, Tech Stacks, and Career Prospects
Deep analysis of three core LLM roles—Application Engineer, Development Engineer, and Algorithm Engineer—covering technical requirements, salary thresholds, and career prospects including RAG, fine-tuning, and inference deployment.
Read more →
Tutorials
DeepSeek V3 + bolt.html: A Practical G…
·2 min
DeepSeek V3 + bolt.html: A Practical Guide to Generating Beautiful Web Pages with Zero Code
Learn how DeepSeek V3-0324 and open-source tool bolt.html combine to generate beautiful HTML pages with zero code using prompt engineering techniques.
Read more →
Tutorials
Why Qwen3 Is the Best Open-Source Mode…
·2 min
Why Qwen3 Is the Best Open-Source Model for MCP Agent Development
Analysis of Qwen3's advantages for MCP agent development, comparing DeepSeek R1's lack of Function Calling, covering MoE architecture and thinking mode switching.
Read more →
Tech Frontiers
June AI Showdown: Mythos, Sonnet 4.8, …
·3 min
June AI Showdown: Mythos, Sonnet 4.8, and GPT-5.6 All Revealed
June 2025 becomes AI's densest release month: Anthropic Mythos nears launch, Claude Sonnet/Opus 4.8 skip-level upgrades, GPT-5.6 rapid iteration, DeepSeek V4 Pro permanent 75% price cut.
Read more →
Tech Frontiers·1 min
DeepSeek V4-Pro Permanent Price Cut: Lower Developer Costs as LLM Price War Heats Up
DeepSeek announces permanent discount pricing for its V4-Pro model. Learn how this impacts developers, V4-Pro's competitive edge, and the latest LLM price war trends.
Read more →
Deep Dives·2 min
AI Agent Development Methodology: A Complete Guide from ReAct to Enterprise-Grade Tech Stack
A deep dive into AI Agent development methodology, from the ReAct theoretical framework to a four-layer enterprise tech stack covering model services, Agent types, LangChain, and production deployment.
Read more →
Tech Frontiers·2 min
GLM5 Architecture Leaked: 745B Parameters, DeepSeek V4 May Launch Quantized Smaller Model First
GLM5 code leak reveals 745B-parameter MoE architecture replicating DeepSeek V3. DeepSeek V4 may launch a 200B quantized model first, with flagship exceeding 1T parameters.
Read more →
Product Reviews·2 min
Kimi K2.6 Open-Source Hands-On: How Strong Is Its Orchestration of 300 Concurrent Agents?
Deep analysis of Moonshot AI's open-source Kimi K2.6 Agent orchestration: 300 sub-Agents executing 4000-step tasks, outperforming GPT-5.4 in coding benchmarks, LoRA fine-tuning on 2x RTX 4090s.
Read more →
Product Reviews·3 min
Kimi K2.6 In-Depth Review: A Complete Breakdown of Its Coding and Agent Capabilities
In-depth review of Kimi K2.6's coding, Agent collaboration, and visual development capabilities. #1 open-source on SWE-Bench Pro, 300 parallel sub-agents, API priced at 1/3 of competitors.
Read more →
Product Reviews
Running Qwen3.6-27B Locally on Mac: 4 …
·3 min
Running Qwen3.6-27B Locally on Mac: 4 Solutions Benchmarked
Benchmarking 4 solutions for running Qwen3.6-27B locally on Mac: GGUF, MLX Diflash, and MTP-LX. MTP-LX 4bit leads at 43.6 tok/s with solid coding, writing, and reasoning quality.
Read more →
Tech Frontiers·2 min
Kimi K2.5 Fully Open-Sourced: Deep Dive into 1T Parameter MoE Architecture + Agent Cluster Capabilities
Deep dive into Moonshot AI's fully open-sourced Kimi K2.5: 1T parameter MoE architecture, Vision-to-Code capabilities, and 100-Agent parallel cluster system topping open-source benchmarks.
Read more →
Tutorials
Universal AI Prompts for Mathematical …
·3 min
Universal AI Prompts for Mathematical Modeling: A Zero-to-Hero Four-Stage Practical Guide
A detailed guide to the four-stage universal AI prompt system for mathematical modeling, covering problem analysis, innovative model construction, data processing, and model solving for competitions.
Read more →