Ivan Rocha

I'm pretty excited to test this one: Gemopus-4-26B-A4B-it-GGUF Q6_K Using @spiritbuun Llama.cpp TurboQuant fork: - Speed: 75 tokens/sec - VRAM usage: 95% (22.7 GB) - Context size: 131072 - GPU: RTX A5000 (Ampere) 24 GB Pretty amazing that you can fit this entire model on GPU with Q6 quality and still have room for a large amount of context! Plus MoE models are still fast at higher quality. Woodchuck Norris vibe check: PASSED Square root of 999999999 -> Correct Hermes Agent -> Interesting behavior. Retains 26B's speed on short prompts, thinks deeply for more complex requests - sometimes thinks a little too much, it might be worth playing with top + temp settings Coding test -> One-shotted a fully working Tetris game - no other MoE model including vanilla 26B was able to do this A very interesting model -m Gemopus-4-26B-A4B-it-Preview-Q6_K.gguf --n-gpu-layers 99 --ctx-size 131072 --cont-batching --cache-type-k turbo4 --cache-type-v turbo4 --fit on --jinja --reasoning-format auto --flash-attn on huggingface.co/Jackrong/Gemop…

🚩GLM-5.1 INT4 model is now available on @huggingface: huggingface.co/INC4AI/GLM-5.1… The model is quantized by AutoRound: github.com/intel/auto-rou…

Tabular Database Systems duckdb.org/library/tabula…

Qwen3.6 Plus vs GPT 5.4 vals.ai/comparison?mod…

SWE-bench Verified and Terminal-Bench—two of the most cited AI benchmarks—can be reward-hacked with simple exploits. Our agent scored 100% on both. It solved 0 tasks. Evaluate the benchmark before it evaluates your agent. If you’re picking models by leaderboard score alone, you’re optimizing for the wrong thing. 🧵

llama.cpp release b8699 trouxe KV cache attention rotation ligada por default. Resultado pratico: Q8_0 fica praticamente lossless (tempo de inferencia sem comprometer qualidade) e o impacto do Q4_0 no KV cache ficou bem menor do que era antes. Traducao pra quem roda modelo local: mais contexto util pela mesma quantidade de VRAM. KV cache quantizado e o multiplicador silencioso que ninguem olha, mas faz diferenca real na pratica.

Jackrong-llm-finetuning-guide An Educational, End-to-End LLM Fine-Tuning Pipeline for Beginners and Developers github.com/R6410418/Jackr…

Qwen 3.6 Plus, Alibaba's latest flagship model, available exclusively through Fireworks! app.fireworks.ai/models/firewor…

（1/8）🚀 Introducing Qwen3.6-Plus: Towards Real-World Agents! 🤖 Today, we’re thrilled to drop a major milestone in our journey toward native multimodal agents. Here is what makes Qwen3.6-Plus a game-changer： 💻 Next-level Agentic Coding: Smarter, faster execution. 👁️ Enhanced Multimodal Vision: Sharper perception & reasoning. 🏆 Top-tier Performance: Maintaining leading general capabilities. 📚 1M Context Window: Available by default via our API. Built on your invaluable feedback from the Qwen3.5 era, we’re laying a rock-solid foundation for real-world devs. Get ready to experience truly transformative ✨ Vibe Coding ✨. Huge thanks to our community! Go try it out and show us what you can build. 👇 Chat: chat.qwen.ai API: modelstudio.console.alibabacloud.com/ap-southeast-1… Blog: qwen.ai/blog?id=qwen3.6 🔔Noted：More Qwen3.6 models to come and be open-sourced! Stay tuned~ 👀#Qwen #AI #AgenticCoding #VibeCoding #Agents

Crush - Your tools, your code, and your workflows, wired into your LLM of choice github.com/charmbracelet/…

Go is simple, so I ended up writing an 865-page book about how it works internally, just to see how it maintains that simplicity 😇

Announcing Amazon S3 Files. The first and only cloud object store with fully-featured, high-performance file system access. Learn more here. go.aws/4tw17Zg

OpenDataLoader PDF PDF Parser for AI-ready data. Automate PDF accessibility github.com/opendataloader…

Benchmark real do Qwopus MoE 35B PolarQuant: PPL: 6.56 (converge pro mesmo nivel do BF16 = 6.54) Velocidade: 37.4 tok/s (2.3x mais rapido que BF16) VRAM: 25GB com cache, 8GB com LRU. CABE NA RTX 4090. De 72GB pra 8GB de VRAM efetiva. Qualidade identica, 2.3x mais rapido. huggingface.co/caiovicentino1…

For two decades, S3 has been an object store, but today it's something broader. S3 Files lets you mount any bucket as a filesystem—no copies, no sync scripts, no choosing between file and object. @andywarfield tells the full story, including the "filerectories" that almost made the cut. allthingsdistributed.com/2026/04/s3-fil…

SOTA on SWE-Bench Pro (58.4): GLM-5.1 delivers significant leaps in coding and agentic performance.

Introducing GLM-5.1: The Next Level of Open Source - Top-Tier Performance: #1 in open source and #3 globally across SWE-Bench Pro, Terminal-Bench, and NL2Repo. - Built for Long-Horizon Tasks: Runs autonomously for 8 hours, refining strategies through thousands of iterations. Blog: z.ai/blog/glm-5.1 Weights: huggingface.co/zai-org/GLM-5.1 API: docs.z.ai/guides/llm/glm… Coding Plan: z.ai/subscribe Coming to chat.z.ai in the next few days.

If you want to get into eBPF programming, I highly recommend Teodor Podobnik's tutorials on iximiuz Labs. The series starts from the basics and goes all the way up to solving practical networking problems. All posts are well-illustrated and full of examples that actually work. Check it out labs.iximiuz.com/tutorials?auth…

🚨 New qwopus model from Jackrong on @huggingface Qwopus3.5-27B-v3-FP8-vllm-ready > uses FP8 quantization, closer to og model > vllm optimized, much faster inference > better quality retention than ggufs huggingface.co/Jackrong/Qwopu…

Merkle trees are everywhere: - ZFS uses them to detect data corruption - Git uses them to verify repo integrity - Cursor uses them for codebase sync - Bitcoin uses them for transaction verification Talked through how they work on the latest database stream.