Open-weight Kimi K2.6 takes on GPT-5.4 and Claude Opus 4.6 with agent swarms

Moonshot AI has released Kimi K2.6 as an open-weight model designed to compete with GPT-5.4 and Claude Opus 4.6 on coding benchmarks. The model can run up to 300 agents in parallel for complex…

Via The Decoder

Kimi 2.6 is now available on @hf.co 🔥🎉
huggingface.co/moonshotai/K...

✨ 1T MoE / 32B active / 256K context
✨ Agent Swarm: 300 sub-agents × 4,000 steps
✨ Modified MIT

New Kimi!!!! It's open weight immediately and it scores really really good on the benchmarks!!! It's gotten even better on long horizon work, seemingly a major focus for Moonshot

www.kimi.com/blog/kimi-k2-6

A bar chart titled "percentage (%)" comparing the performance of four AI models across various benchmarks. The models are: **Kimi K2.6** (blue), **GPT-5.4 (xhigh)** (white), **Claude Opus 4.6 (max effort)** (light gray), and **Gemini 3.1 Pro (thinking high)** (white). The benchmarks are divided into three categories: ### General Agents * **Humanity’s Last Exam (Full) w/ tools:** Kimi K2.6 leads with 54.0. * **BrowseComp:** Gemini 3.1 Pro leads with 85.9; Kimi K2.6 follows at 83.2. * **DeepSearchQA (f1-score):** Kimi K2.6 leads significantly with 92.5. * **Toolathlon:** GPT-5.4 leads with 54.6; Kimi K2.6 is at 50.0. * **OSWorld-Verified:** GPT-5.4 leads with 75.0; Kimi K2.6 is at 73.1. (Claude is 72.7, Gemini is missing data). ### Coding * **Terminal-Bench 2.0 (Terminus-2):** Gemini 3.1 Pro leads with 68.5; Kimi K2.6 is at 66.7. * **SWE-Bench Pro:** Kimi K2.6 leads with 58.8. * **SWE-bench Multilingual:** Claude Opus 4.6 leads with 77.8; Kimi K2.6 is at 76.7. (Gemini is 76.9, GPT is missing data). ### Visual Agents * **MathVision w/ python:** GPT-5.4 leads with 96.1; Kimi K2.6 is at 93.2. * **V* w/ python:** GPT-5.4 leads with 98.4; Kimi K2.6 and Gemini 3.1 Pro are tied at 96.9.

Kimi 2.6: Hangs with the best

* on par with Opus 4.6 & GPT-5.4 5.4 xhigh
* long horizon coding tasks

www.kimi.com/blog/kimi-k2-6

If your classification/extraction needs to be reliable, order our AI-driven spreadsheet processing services for classification/extraction tasks with Perplexity AI or Ollama LLM models integration.
matasoft.hr/qtrendcontro...
#AI #SpreadsheetAI #DataAnalysis #DataExtraction #DataCategorization

If your “free text” description field contains multiple facts in unorganized manner, Matasoft can extract and split it into separate columns you can actually use. Try our AI-driven spreadsheet services! matasoft.hr/qtrendcontro...
#AI #SpreadsheetAI #DataAnalysis #DataExtraction #DataCategorization

There's a new AI-powered spreadsheet in town. (Un)Perplexed Spready handles unstructured text, formatting, and classification with AI-driven formulas with advanced LLM prompts. #Spreadsheet #DataAnalysis #Tools #AI #LLM #FutureOfWork #SpreadsheetAI #AITools #DataAnalysis matasoft.hr/QTrendContro...

Reasoning Meets Instruction: Exploring DeepHermes-3 8B for the Ascend Ecosystem

At only 8B parameters, DeepHermes-3 retains coherent chain-of-thought reasoning rarely seen in compact models. Optimized for Huawei Ascend NPUs via GGUF, it's a standout for local agentic workflows where hardware constraints matter. A rare blend of DeepSeek's logic and Nous Hermes'...

Hermes Agent with Ollama: The Self-Improving AI Agent That Runs Entirely Local Hermes Agent from Nous Research runs entirely local through Ollama with 70+ built-in skills, cross-session memory, and messaging gateway integrations. Setup guide, model benchmarks, and comparison with Claude Code for autonomous workflows.

Hermes Agent with Ollama: The Self-Improving AI Agent That Runs Entirely Local

Hermes Agent from Nous Research runs entirely local through Ollama with 70+ built-in skills, cross-session memory, and messaging gate…

#AI #Agents #Hermes

pooya.blog/blog/hermes-agent-ollama...

8234b71d-394d-4d8f-94bd-3fe5668af346

1/ Hermes Agent by Nous Research is the first AI with a built-in learning loop. It creates skills from experience and gets smarter the longer it runs.

#ai #llm #futureofwork #spreadsheetai #aitools #dataanalysis | Matasoft - Straight-forward IT Solutions What if your spreadsheet could think? (Un)Perplexed Spready integrates LLMs like GLM-5.1 directly into cells. Automate complex tasks from taxonomy categorization or pdf data extraction to web research...

Spreadsheets got smarter with AI. Discover (Un)Perplexed Spready: AI-embedded formulas that classify data, extract info, and standardize text automatically. Transform your workflow.
www.linkedin.com/posts/mataso...
#AI #LLM #FutureOfWork #SpreadsheetAI #AITools #DataAnalysis

Screenshot of the repository

Self-improving AI agent with built-in learning loop

Turn raw datasets into analytics-ready tables (structured columns + consistent labels) without building a data pipeline. We can perform data extraction and categorization with power of artificial intelligence, embedded into spreadsheet formulas. matasoft.hr/qtrendcontro...
#AI #Spreadsheets

Get repetitive and tedious spreadsheet work off your plate: extraction + categorization + labeling delivered as a finished spreadsheet. Try our AI-driven spreadsheet processing services!
matasoft.hr/qtrendcontro...
#AI #SpreadsheetAI #DataAnalysis #DataExtraction

Nous Research Launches Hermes Agent - An Open Source Agent That Remembers Everything Nous Research's Hermes Agent is an open-source CLI agent with persistent multi-level memory, cross-platform messaging support, subagent delegation, and a growing skills ecosystem.

Nous Research Launches Hermes Agent - An Open Source Agent That Remembers Everything

awesomeagents.ai/news/nous-research-herme...

#NousResearch #AiAgents #OpenSource

Hermes Agent is a self-improving AI agent that builds skills from experience, uses FTS5 session search, spawns parallel subagents, and supports Telegram/Discord/CLI frontends. Model-agnostic and serverless-ready. #tool #ai https://bit.ly/3Oa0rtM

OpenClaw vs. Hermes Agent: The race to build AI assistants that never forget OpenClaw and Hermes Agent take different approaches to persistent AI coding assistants. One prioritizes ecosystem reach, the other deep learning over time.

OpenClaw and Hermes Agent take different approaches to persistent AI coding assistants. One prioritizes ecosystem reach, the other deep learning over time.

The Hermes agent and Karpathy's LLM wiki have been getting a lot of attention lately (as they should). But even before that, I’d been building a repo-level #knowledge management system. It should work with any #agent and any tool. I still like to work in #code in VS Code.

github.com/Hypercubed/A...

📝 Summary:

Hermes Agent is a self-improving AI assistant from Nous Research that runs across multiple platforms (CLI and messaging gateways), supports a wide range of models/tools, and includes a built-in learning loop, memory, and scheduling capabilities. It offers extensive documentation, (1/2)

So where are all the AI apps? – Answer.AI Practical AI R&D

Fantastic post answering a question I’ve had, if we can all write all the code now, where are all the cool new code bases and tools?

www.answer.ai/posts/2026-0...

If you have thousands of rows of semi-structured text, outsource extraction into clean structured fields. We can perform data extraction and categorization with power of artificial intelligence, embedded into spreadsheet formulas.
matasoft.hr/qtrendcontro...
#Automation #AI #DataGovernance #LLM #MDM

Turn customer feedback into actionable buckets (shipping/quality/support/etc.) with proper tagging, data extraction and classification in the spreadsheet. Try our AI-driven spreadsheet processing services!
matasoft.hr/qtrendcontro...

The spreadsheet is dead. Long live the intelligent spreadsheet. 2,751 products can't be wrong: matasoft.hr/QTrendContro... #FutureOfWork #Automation #AI #DataGovernance #LLM #ProductivityTools #UnPerplexedSpready

Spreadsheets just got smarter. Discover (Un)Perplexed Spready: AI-embedded formulas that classify data, extract info, and standardize text automatically. Transform your workflow. #Productivity #Tech #Innovation #AI #LLM #SpreadsheetAI matasoft.hr/QTrendContro...

Real businesses run on messy text columns. (Un)Perplexed Spready is built for tasks like extracting attributes, standardizing names/addresses, and categorizing products—without leaving the spreadsheet. matasoft.hr/QTrendContro...
#DataQuality #AI #LLM #DataClassification #Ecommerce #Spreadsheets

Practical AI in spreadsheets: feed product label + internal taxonomy + target taxonomy list into a single formula call, and let the model return the best-fit categories (including multiple matches when needed). matasoft.hr/QTrendContro... #DataClassification #Ecommerce #Spreadsheets #AI

Quote Tweet: https://twitter.com/i/status/2044768734234243427

Qwen 3.6 is here, and open-source! Run it locally with improved agentic coding capabilities.

Try it with Claude Code:
ollama launch claude --model qwen3.6

Try it with OpenClaw:
ollama launch openclaw --model qwen3.6

Run it:
ollama run qwen3.6

Qwen3.6 35B-A3B can now be run locally! 💜

The model is the strongest mid-sized LLM on nearly all benchmarks.

Run on 23GB RAM via Unsloth Dynamic GGUFs.

GGUFs to run: huggingface.co/unsloth/Qwen...
Guide: unsloth.ai/docs/models/...

Qwen 3.6 Ships a 35B MoE That Codes Like Models 10x Its Size Alibaba's Qwen 3.6-35B-A3B activates only 3B of its 35B parameters per token, scores 73.4% on SWE-bench Verified, handles video and images, and ships under Apache 2.0.

Qwen 3.6 Ships a 35B MoE That Codes Like Models 10x Its Size

awesomeagents.ai/news/qwen36-35b-a3b-agen...

#Qwen #Alibaba #OpenSource

If your catalog has an internal category tree but the marketplace needs a different taxonomy, this workflow shows how to map at scale using AI formulas + a lookup taxonomy sheet. matasoft.hr/QTrendContro...
#ProductData #MasterData #AI