#throughput
680 related articles
·2 min
DeepSeek V4 Flash Free Access: Complete Breakdown of the 0731 Version on InferX
InferX offers free access to DeepSeek V4 Flash (0731 version) with zero data retention and OpenAI-compatible API. Full breakdown of features, pricing, and developer value.
Read more →
·2 min
A Slow-Mode Vision for Codex: Doubling Coding Quota with Flex API
A developer proposes a Flex API-based slow mode for Codex, trading speed for nearly double the usage quota. We analyze the product logic, technical feasibility, and business challenges.
Read more →
·3 min
How Cheap Can It Be to Serve a 2.8 Trillion Parameter Model? A Deep Dive into Inference Economics
Deep analysis of the real cost of serving a 2.8 trillion parameter model. From MoE sparse activation to batching scale effects and inference optimization, revealing why model size and serving cost are less correlated than assumed.
Read more →
·2 min
The AI Model Value-for-Money Battle: How to Rationally Choose the Right LLM for You
As AI LLM capabilities converge, cost-effectiveness becomes the key selection factor. This article explores how to rationally compare AI models through value assessment, task matching, and cost-benefit analysis.
Read more →
·3 min
You Don't Always Need a Vector Database for Vector Search — Brute Force Might Be All You Need
In-depth analysis of when brute force vector search beats vector databases. For RAG apps with under a few hundred thousand vectors, brute force offers exact recall, simpler architecture, and easier debugging.
Read more →
·4 min
Heretic Uncensoring Technology Explained: Practical Applications on Jamba2 and Qwen3.5 Open-Source Models
Deep dive into Heretic uncensoring technology applied to Jamba2-Mini, Qwen3.5-9B, and 27B open-source models, exploring how refusal rates dropped from 97% to 4% and the safety debates involved.
Read more →
·5 min
The Complete Learning Path to Understanding the Kimi K3 Technical Report
A systematic learning path for understanding the Kimi K3 technical report, covering MoE, MLA, distributed training, and modern post-training techniques.
Read more →
·3 min
James Kuffner Deep Dive: From the RRT Algorithm to Symbotic's Large-Scale Warehouse Robotics
Explore RRT co-inventor James Kuffner's career from Cloud Robotics and Google Robotics to Symbotic CTO, driving robots from labs to Walmart warehouse-scale deployment.
Read more →
·3 min
How GitHub Achieves 45 GiB/s Single-Core Case-Folding: An In-Depth Optimization Analysis
Deep dive into how GitHub achieves over 45 GiB/s single-core case-folding using branch-free loops, byte-space arithmetic, and SIMD vectorization, approaching memory bandwidth limits.
Read more →
·2 min
GPT 5.6 Luna: Outperforms Google's Best Model at a Price Lower Than Google's Cheapest
GPT 5.6 Luna reportedly tops Google's flagship on the Artificial Analysis Intelligence Index while costing less than Google's cheapest model. A deep dive into the tech trends, industry impact, and developer implications.
Read more →
·2 min
DeepSeek-V4-Flash Model Analysis: A New Lightweight AI Choice Prioritizing Efficiency
In-depth analysis of DeepSeek-V4-Flash model's product positioning and technical approach. Examining lightweight trends through the Flash naming, MLA attention mechanism, MoE architecture evolution, and implications for the open-source AI ecosystem.
Read more →
·2 min
DeepSeek-V4-Flash Model Analysis: A New Lightweight AI Choice Prioritizing Efficiency
In-depth analysis of DeepSeek-V4-Flash model's positioning and technical path. Exploring the lightweight trend behind the Flash naming, MLA attention, MoE architecture, and its significance for open-source AI.
Read more →
·3 min
DeepSeek V4 Flash Analysis: The Three-Dimensional Balance of Intelligence, Performance, and Price
Deep analysis of DeepSeek V4 Flash 0731 across intelligence, performance, and price dimensions, exploring how this Chinese LLM delivers extreme cost-performance to reshape the AI industry.
Read more →
·3 min
Fix for Immich Crashes Caused by Proxmox's Default kvm64 CPU Type
Proxmox's default kvm64 CPU type only exposes x86-64-v1 instructions, causing Immich's ML container to crash when NumPy's baseline isn't met. Learn the root cause and one-command fix.
Read more →
·3 min
Supabase in Production: A Deep Dive into Cron and Queue Reliability Issues
Deep analysis of Supabase pg_cron and pgmq reliability issues in production, including task loss, execution uncertainty, and observability gaps, with practical architecture optimization advice.
Read more →
·3 min
Supabase in Production: A Deep Dive into Cron and Queue Reliability Issues
Deep analysis of Supabase production reliability issues with pg_cron and pgmq, covering task loss, execution uncertainty, and observability gaps, with practical architecture optimization advice.
Read more →
·3 min
GPT-5.6 Price Cut by 80%: The Deeper Game Behind the AI Model Price War
OpenAI announces GPT-5.6 Luna 80% cheaper, Terra 20% cheaper, targeting Gemini Flash. Analysis of the AI price war's business logic, Google's moat, Microsoft's hidden gains, and open-source impact.
Read more →
·2 min
GPT-5.6 Released: Redefining the Price-Performance Boundary of AI Models
OpenAI releases GPT-5.6, targeting the price-performance frontier. Analysis of how architectural optimization and inference efficiency reduce costs, and how LLM competition shifts from capability to cost efficiency.
Read more →
·5 min
pgvector 0.8 Iterative Index Scans in Practice: Best Practices for Vector Search with Scalar Filtering
Deep dive into pgvector 0.8 iterative index scans that solve the over-filtering problem in hybrid vector search with WHERE conditions. Includes SQL examples, strict_order vs relaxed_order comparison, and max_scan_tuples tuning tips.
Read more →
·6 min
Securely Exposing Self-Hosted Services via VPS: Architecture & Practical Guide
Learn how to securely expose self-hosted services via VPS using WireGuard tunnels, Pangolin, and reverse proxies — enabling client-free access to Plex, Immich, and more while hiding your home IP.
Read more →