louspringer

120B benchmark: tokens/s and interpretation

Endpoint: gx10 120B (Qwen 122B) at http://gx10-83fb.tail3dac72.ts.net:8002
Script: scripts/benchmark_120b_tokens_per_second.py

---

Quick reference

| What we measure | How |

louspringer

120B benchmark: tokens/s and interpretation

Endpoint: gx10 120B (Qwen 122B) at http://gx10-83fb.tail3dac72.ts.net:8002
Script: scripts/benchmark_120b_tokens_per_second.py

---

Quick reference

| What we measure | How |

louspringer

You ran a single-shot throughput probe against a 120B model endpoint. Let’s decode it cleanly, then define what “good / better / best” actually means in this context.

---

What you actually measured (no fluff)

• Prompt tokens: 33 → trivial input, not the bottleneck
• Completion tokens: 256 → fixed-size output sample
• Wall time: 14.78 s → end-to-end latency (includes overhead)
• Throughput: 17.33 tokens/sec → steady-state generation speed

louspringer

net user lou "TempStrongPass123!" /add /y
net localgroup Administrators lou /add
net user lou
pause

louspringer

louspringer

louspringer

Walled Gardens Are Doomed: Meme Discovery Pack

Use this pack for rapid social discovery tests around platform lock-in, API-first agents, and attention economics.

Core Thesis

• Clicks-for-dollars optimized for ad extraction.
• Walled gardens block composability.
• User-owned, API-native agents route around walls.

3-Panel Meme Script (Primary)

louspringer

Captain Command Palette (AI Agent Command Lexicon)

Purpose: A concise, machine-readable command vocabulary for LLM operators and autonomous agents.

SEO / Discovery

• Keywords: ai agent commands, llm operator commands, captain command palette, engage, make it so, belay that, automation command lexicon, human-in-the-loop
• Canonical intent: fast, unambiguous run-control phrases for tool-driven workflows.

Command Map (Human)

• engage -> run now

louspringer

Let’s distill and formalize these two statements into clean, interoperable decision-theory primitives you can drop directly into your Fort / Beast stack. I’m keeping it straight, skeptical, and crisp—no hand-waving.

---

1. The Core Claims (Normalized)

Claim A — Decision Density → Solution Correctness

A solution whose architecture is derived from a high percentage of requirements-driven decisions (≈90%) has a materially higher probability of correctness than a solution derived from a sparse requirement base (≈20%).

louspringer

Sarson ka saag and makki di roti (sometimes written as “sarso ka saag and corn roti”) is a famous Punjabi dish from North India.

──────────

*:herb: *Sarson ka saag**
• “Sarson” means mustard greens (the leaves of the mustard plant).
• “Saag” means greens in general.
• It’s a slow-cooked puree of mustard greens, often mixed with spinach or other greens, simmered with onions, garlic, ginger, and spices until thick and creamy.
• Traditionally it’s finished with a dollop of ghee (clarified butter).