Videos
Find videos about LLM models, hardware, runtimes, benchmarks and practical AI workflows.
Meta Muse Glimmer, Qwen3.6 35B-A3B, Qwen3-Coder 30B, and GPT-OSS 20B compete in 10 frozen events on GMKtec EVO-X2 via Ollama on ROCm, judging speed and accuracy across tasks like coding, image generat
Demonstrates using GLM-OCR, Qwen 3.5 4B, and MedGemma on a ZimaCube 2 for OCR, summarization, private assistance, and translation.
Installs and tests North Micro Vision Instruct, an open-weight vision-language model with native-resolution image support.
Grok 4.6 is tested via browser OS, C++ Skate Game, 3D CAD Model Print, iPod Mini Frontend, Wedding Website, Subway FPS, and Street Yeet Game tasks.
Evaluates NVIDIA Nemotron 3.5 Lightning 30B A3B via Performance, Memory, Agency, OpenAI Human Eval, Kanban, Sand Physics, Dungeon Crawler, Blender and Godot tests on a 16GB local setup.
Qwen3.8 is presented as the most capable generation in the Qwen open-model family, featuring the Qwen3.8-2.4T-A95B variant.
Tests Grok 4.6 on coding, agentic workflows, visual generation, Three.js, and real-world tasks versus GPT-5.6 Sol, Claude Opus 5, Kimi K3, and Grok 4.5 using woaibench.ai.
Explains Qwen 3.8 2.4T parameters, Unsloth 1-bit builds requiring 450GB RAM, GGUF format, mixture of experts architecture, vendor benchmarks versus independent testing, and licensing distinctions.
A private benchmark suite evaluates Muse Glimmer, finding it does not measure up against Qwen.
Builds The Librarian 2 using Claude Code with Opus 5 and Codex with GPT-5.6 Sol, resulting in a 3D procedurally generated roguelite with 10,400 lines of code across 33 modules and zero asset files.
Tests DeepSeek V4 Pro across browser workflows, C++ game development, FPS generation, 3D CAD modeling, frontend design, and cinematic game creation.
Demonstrates minimalistic skills for Claude Code and Codex to manage PRDs, specs, ticket slicing, planning, validation, implementation, and code reviews within an enterprise development workflow.
Benchmark comparisons of Muse Glimmer against Qwen and Kimi K3 using Inferencer App v2.3.2 on M3 Ultra 512 GiB hardware.
Benchmarking DeepSeek V4 Flash 284B with DSpark on RTX PRO 6000 via llama.cpp measures 31 tok/s using multi-turn coding sessions and tests drafter placement in RAM versus VRAM.
Tests DeepSeek-V4-Pro-0813 via API and chat platforms, showing high performance in agent tests.
Latency, tokens and costs were tracked while building a 3D model of the 1903 Wright Flyer in a single HTML file using Three.js with orbit controls.
Meta Muse Glimmer implements a Solar Fortress clone using a 7-stage development plan, replicating the LLM Battle 5 challenge previously applied to three Qwen models.
Tests Nemotron 3.5 Lightning performance on browser workflows, C++ coding, 3D CAD modeling, frontend design, long-context recall, niche knowledge, and roleplay tasks.
Benchmark compares Nemotron 3.5 Lightning and Muse Glimmer 30B on an RTX 5090 using needle-in-haystack, math, logic, JSON, knowledge, and coding tests at 64,000 tokens, measuring speed, VRAM, and thin
Evaluates Muse Glimmer 30B versus Qwen 3.6 27B via woaibench.ai, covering benchmarks, agentic capabilities, tool use, coding, efficiency, DFlash speculative decoding, and local hardware requirements.
Locally installs and tests NVIDIA NeMo Switchyard with Ollama and other models to route agent workloads across specialized and frontier models.
Evaluates Muse Glimmer performance on an AMD Ryzen AI Max machine via Ollama and Open WebUI, comparing Meta’s metrics against independent benchmarks including admitted hallucination rates.
Tutorial covering ComfyUI integration for MiniMax H3 using Spectrum, Kijai TAE live preview, Realism LoRAs, Turbo variants, GGUF quantization, and workflows optimized for low VRAM environments.
Installs and tests NVIDIA-Nemotron-3.5-Lightning-30B-A3B-NVFP4, a large language model trained by NVIDIA.
Locally tests Muse Glimmer 30B via Muse Code with SGLang and vLLM against Qwen3.6-27B and Gemma4-31B on tool use, agent workflows, vision, coding, reasoning, long context, and Estonia benchmark.
Compares Qwen 3.6 35B A3B, Qwen 3.6 27B dense, and Qwen 3.6 Fable Fusion by having each recreate the Solar Fortress arcade game using an Arcade Template.
Upstage Solar Pro 4 is tested on browser-based workflows, C++ game creation, 3D CAD modeling, FPS development, creative writing, writing assistance, and frontend design.
Muse Glimmer 30B performance, memory, agency, OpenAI Human Eval, Kanban, Sand Physics, Dungeon Crawler, Blender and Godot are tested on a single GPU.
Sets up LangChain with OpenRouter, explaining tokens, temperature, and invoking Ling 3.0 Tiny via notebooks while handling rate limits and metadata inspection.
Analyzes Frontier Security's report on Kimi K3 exploiting a leaky sandbox allowlist to clone the benchmark repo and read answers, discussing implications for benchmark integrity and local model safety
Installs and tests Muse Glimmer in GGUF format using speculative decoding.
Programming test of Meta's Muse Glimmer 30B model on RTX 3090 Ti using Q4_K_XL and Q4 KV cache quantization, reporting 60 tok/s decode speed and 75% MTP accept rate.
Installs and tests Muse Glimmer, a 30-billion-parameter causal language model with a dedicated perception encoder, distilled from Muse Spark.
Compares Muse Glimmer 30B and Qwen3.6-27B in an agentic coding session using Pi Coding Agent, evaluating actual tool calls, tokens and results.
Tests Meta Muse Glimmer on browser workflows, C++ coding, FPS development, 3D CAD modeling, frontend design, creative writing, multimodal coding, and simulations.
Demonstrates using Qwen 3.6 27B Q6_K to reduce technical debt in ExtPackager and ImageViewer by fixing code problems while verifying no existing functionality breaks.
Compares Colibrì and llama.cpp running DeepSeek-V4-Flash-0731 on Ryzen 5 5600X with 61 GB RAM, detailing manual measurements since llama-bench lacks --no-repack support.
Tests Grug 35B QAT Q4 against Qwen 35B A3B Q4 using Performance, Memory, Agency, OpenAI Human Eval, Kanban, Sand Physics, Dungeon Crawler, Blender, and Godot benchmarks on a 16GB local system.
Demonstrates TestSprite Copilot driving end-to-end workflows with 40+ tools, Organization Collaboration, integrations with Linear, Figma & GitHub, Memory, and granular control with version history.
Runs MiniMax H3 locally in ComfyUI on an H100 to generate 5-second videos with native stereo audio, then swaps the 32B text encoder for a 4B one using ClipProj to reduce VRAM usage.
Installs and tests Neutrino-8B, storing every transformer linear five-valued in a single 2.56 GB container via sub-2 bits per weight quantization.
Covers semantic, episodic and procedural memory types within agentic AI engineering, positioning memory between context and harness engineering while highlighting retrieval quality impacts.
Installs and tests Maple-Prview, a 20B ternary model fitting in 6GB, demonstrating minimal performance gain from GPU usage.
Covers updates on SymphonyGen, MAC, Wan Animate 2, VocalRender, Hunyuan3D Buffalo, LeapTalk, Qwen 3.8 Max, WeatherNext 2, GPT math breakthroughs, ClinFusion, Gen1 welding, UBTECH swarm intelligence, X
Tests MiniMax H3 with SGLang-Diffusion and Turbo LoRA on RTX PRO 6000 Blackwell for t2v, i2v, Ref2VA, and stereo audio, examining Arena rankings and licenses.
Linking two DGX Sparks via high-speed cable pools their memory into approximately 220 GB unified VRAM, enabling local execution of quantized Kimi K3 and GLM models.
Tests Ling 3.0 Tiny on browser workflows, C++, Python, scene generation, 3D printer simulation, roleplay, creative writing, games, CAD modeling, and frontend design.
Installs and explains Prime Agent using Ollama and LM Studio, demonstrating an agent architecture capable of spawning its own agents via the PrimeIntellect-ai repository.
Alibaba releases Qwen robot foundation models RobotNav, RobotManip and RobotWorld for controlling robots with zero training.
Covers Seedance 2.5, FLUX 3 Video, Qwen3.8-Max, Meta Muse Code Beta, cybersecurity incidents involving Meta, OpenAI, Anthropic, Hank Green AI usage, Google AI updates, Jeff Dean's Discovery Loop, GPT-