Tag

#ai

15 posts tagged #ai. Show all.

Posts

15 posts

Jul 17, 2026 A new fair metric for comparing LLMs by intelligence and cost
A new way to measure LLM value as a combination of intelligence and price - fitted to the market, resistant to cheap-model dominance, and visualized live in ollamadash.
ai metrics ollamadash
Jun 29, 2026 Model analysis summary
A deep dive into the different parts of the model ids - parameters, quantization, but also architectural stems like MXFP8, NVFP4, MLX, etc. - what they mean and how they interact with each other. Semi-raw analysis from claude code.
ai llm quantization
Jun 29, 2026 Model-ID stem analysis - quality / speed / size rankings and combinability
Analysis of the architectural and quantization stems in model identifiers, ranking them on quality, speed (TTFT and TPS), and size across Mac, Linux, and WSL2 - plus a combination matrix of what stacks and what is mutually exclusive.
ai llm quantization bibliography
Jun 29, 2026 Model-ID glossary research notes
Verified findings and source URLs collected as encountered during research for the model-ID glossary - covering GGUF quantization, NVFP4/MX formats, MLX, QAT, MTP, MoE notation, YaRN, distillation, MatFormer, DPO, LASER, reasoning models...
ai llm reference
Jun 29, 2026 Model glossary run 01 - Core ID inventory
Run 01 of the model-ID glossary analysis - extracts and counts identifier strings, families, quantization labels, variants, parameter sizes, and providers across the Ollama, OpenRouter, and Artificial Analysis datasets.
ai llm reference
Jun 29, 2026 Model glossary run 02 - Unified stem frequency table
Run 02 of the model-ID glossary analysis - tokenizes every identifier across all three datasets and produces a unified stem frequency table, including special architectural tokens.
ai llm reference
Jun 29, 2026 Model glossary run 03 - Advanced quant/format/training stems
Run 03 of the model-ID glossary analysis - lists Ollama tags carrying advanced quantization, format, and training stems (NVFP4, MXFP8, FP8, QAT, MTP, MLX, INT4, INT8).
ai llm reference
Jun 29, 2026 Model glossary run 04 - Codename context resolution
Run 04 of the model-ID glossary analysis - resolves context for obscure codename tokens (rnj, hy3, jt, rsnsft, laguna, ling, etc.) by matching them against identifier strings across all three datasets.
ai llm reference
Jun 29, 2026 Model glossary run 05 - Performance & combination research notes
Run 05 of the model-ID glossary analysis - verified findings and source links on the GGUF quant ladder, MLX vs llama.cpp, NVFP4/MXFP4/FP8, QAT, TTFT vs TPS, and orthogonal build-time techniques.
ai llm quantization
Jun 29, 2026 Model glossary run 06 - Stem axis classification and combinability
Run 06 of the model-ID glossary analysis - classifies stems into the three orthogonal axes (numeric encoding, runtime, build-time technique) and tabulates which combinations stack, warn, or are mutually exclusive.
ai llm quantization
Jun 29, 2026 Model glossary run 07 - v2 verification
Run 07 of the model-ID glossary analysis - re-researches the four hedged claims in the analysis report (Ollama MLX backend, FP4 in llama.cpp, Apple Silicon FP4/FP8 hardware, MLX long-context decode) with sources.
ai llm quantization
Jun 29, 2026 Model-ID glossary - every part in model identifiers
A reference that decodes every meaningful stem found in model identifiers across Artificial Analysis, Ollama, and OpenRouter - parameter sizes, quantization/precision formats, runtimes, build-time techniques, capability suffixes, context...
ai llm reference bibliography
Jun 21, 2026 AI learning resources
My current recommendations for learning modern AI - changes over time
ai learning curriculum
Jun 21, 2026 Reposting - trust vs provenance
Reposting a repost from a more credible source adds weight to inner post
ai provenance credibility trust
Jun 19, 2026 My AI stack
AI keeps changing, and the best model, agent, tool, etc. are ever changing - this post retains my current AI and software toolchain, with a log of updates to it over time.
ai updates