shaltielshmid

name	always-concise
description	Enforce concise, clear, ADHD-friendly responses with minimal prose while preserving all necessary information. Use for every response and every task, regardless of topic, unless a higher-priority instruction requires a different format or the user explicitly requests more detail.

Always Concise

Write the shortest complete answer that fully solves the user's request.

These are synthetic testsrc2 clips for sign/simple-video-utils. No third-party footage is included.

Restore the three clips and run the reproducer:

for file in *.mp4.b64; do base64 -d "$file" > "${file%.b64}"; done

	MODEL_TO_USE = "NanoV2" # or "MistralNemo"
	IMPORT_MODEL_AND_DATA = False # set to True to import the model from HF hub, only needs to be run once
	REASONING_STYLE = "gpt_oss_120b" # or DeepSeek_R1_0528
	NUM_NODES = 2
	GPUS_PER_NODE = 8

	import nemo_run as run
	from nemo.collections import llm
	from nemo.collections.llm.gpt.model.mistral import MistralModel, MistralNeMoConfig12B
	from nemo.collections.llm.gpt.model.ssm import MambaModel, NemotronNano12Bv2

	#:package ARSoft.Tools.Net@3.6.1
	#:package LLMTornado@3.7.25

	using ARSoft.Tools.Net;
	using ARSoft.Tools.Net.Dns;
	using LlmTornado.Code;
	using LlmTornado;
	using LlmTornado.Chat;

	(string EnvKey, LLmProviders Provider)[] Providers = [