nowait-reasoning-optimizer
Implements the NOWAIT technique for efficient reasoning in R1-style LLMs
Install / Use
npx skills add foryourhealth111-pixel/Vibe-Skills --skill nowait-reasoning-optimizerInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
AI & Machine LearningSupported Platforms
Tags
Our assessment of nowait-reasoning-optimizer
nowait-reasoning-optimizer scores 88/100 on our quality scale, 447th of 950 AI & Machine Learning skills we index (top 48%).
Its SKILL.md is 4.8 KB long, well organised into 19 sections with 5 code examples: a solid amount of guidance for an agent.
With 3,433 GitHub stars, it is one of the more widely adopted skills in the catalogue.
Maintenance, license and trust
- The repository was last updated about 3 months ago, so nowait-reasoning-optimizer is actively maintained.
- Our last check on 2026-10-31 found the source still online.
- It is released under the Apache-2.0 license, a permissive license that allows use, modification and commercial use with attribution.
- Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
No issues foundOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.
AI review by kimi-k2.7-code on 2026-09-30. Automated pattern scan on 2026-09-30. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
nowait-reasoning-optimizer compared with similar skills
All 4 of these similar skills score higher than nowait-reasoning-optimizer; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| nowait-reasoning-optimizer (this skill)by foryourhealth111-pixel | 88 | 3.4k | 3mo ago | SKILL.md |
| claude-memby thedotmack | 100 | 117.6k | today | CLAUDE.md |
| Understand-Anythingby Egonex-AI | 100 | 92.8k | 1d ago | CLAUDE.md |
| headroomby headroomlabs-ai | 100 | 79.8k | today | CLAUDE.md |
| CowAgentby zhayujie | 100 | 48.0k | today | CLAUDE.md |
Frequently asked questions
- How do I install nowait-reasoning-optimizer?
- Run
npx skills add foryourhealth111-pixel/Vibe-Skills --skill nowait-reasoning-optimizer. The install tabs above show the steps for each supported agent. - Which AI agents does nowait-reasoning-optimizer work with?
- It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is nowait-reasoning-optimizer safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It is Apache-2.0-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is nowait-reasoning-optimizer still maintained?
- The repository was last updated about 3 months ago, so nowait-reasoning-optimizer is actively maintained.
Skill content
View source on GitHubname: nowait-reasoning-optimizer description: Implements the NOWAIT technique for efficient reasoning in R1-style LLMs. Use when optimizing inference of reasoning models (QwQ, DeepSeek-R1, Phi4-Reasoning, Qwen3, Kimi-VL, QvQ), reducing chain-of-thought token usage by 27-51% while preserving accuracy. Triggers on "optimize reasoning", "reduce thinking tokens", "efficient inference", "suppress reflection tokens", or when working with verbose CoT outputs.
NOWAIT Reasoning Optimizer
Implements the NOWAIT technique from the paper "Wait, We Don't Need to 'Wait'! Removing Thinking Tokens Improves Reasoning Efficiency" (Wang et al., 2025).
Overview
NOWAIT is a training-free inference-time intervention that suppresses self-reflection tokens (e.g., "Wait", "Hmm", "Alternatively") during generation, reducing chain-of-thought (CoT) trajectory length by 27-51% without compromising model utility.
When to Use
- Deploying R1-style reasoning models with limited compute
- Reducing inference latency for production systems
- Optimizing token costs for reasoning tasks
- Working with verbose CoT outputs that need streamlining
Supported Models
| Model Series | Type | Token Reduction | |--------------|------|-----------------| | QwQ-32B | RL-based | 16-31% | | Phi4-Reasoning-Plus | RL-based | 23-28% | | Qwen3-32B | RL-based | 13-16% | | Kimi-VL-A3B | Multimodal | 40-60% | | QvQ-72B-Preview | Multimodal | 20-30% |
Important: NOWAIT works best with RL-based models. Distilled models (Qwen3-4B/8B/14B) show degraded performance when reflection tokens are suppressed.
Quick Start
1. Basic Implementation
from scripts.nowait_processor import NOWAITLogitProcessor
# Initialize processor for your model's tokenizer
processor = NOWAITLogitProcessor(tokenizer)
# Use during generation
outputs = model.generate(
inputs,
logits_processor=[processor],
max_new_tokens=32768
)
2. Keywords Suppressed
See references/keywords.md for the complete list. Core keywords:
wait, alternatively, hmm, but, however, check,
double-check, maybe, verify, again, oh, ah
How It Works
- Initialize Keywords: Identify reflection keywords from empirical analysis
- Expand to Token Variants: Map keywords to all token variants in vocabulary (e.g., "wait" → " wait", "Wait", " Wait", ".wait", "WAIT")
- Suppress During Inference: Set logits of reflection tokens to large negative values during decoding
Logits (Before) Logits (After)
Wait 0.8 → Wait -inf
First 0.6 → First 0.6
Hmm 0.5 → Hmm -inf
Let 0.4 → Let 0.4
Key Findings
Why It Works
- NOWAIT doesn't eliminate self-reflection entirely—it guides models to skip unnecessary "waiting" reasoning
- Models still perform essential verification at key decision points
- Results in more linear, straightforward reasoning paths
RL vs Distilled Models
| Model Type | NOWAIT Effect | Recommendation | |------------|---------------|----------------| | RL-based (QwQ, Phi4, Qwen3-32B) | Stable accuracy, significant token reduction | ✅ Recommended | | Distilled (Qwen3-4B/8B/14B) | Accuracy degradation on hard tasks | ⚠️ Use with caution |
Distilled models rely heavily on CoT structure from training data—removing reflection tokens disrupts their reasoning patterns.
Integration Examples
HuggingFace Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
from scripts.nowait_processor import NOWAITLogitProcessor
model = AutoModelForCausalLM.from_pretrained("Qwen/QwQ-32B")
tokenizer = Auto…[redacted]("Qwen/QwQ-32B")
processor = NOWAITLogitProcessor(tokenizer)
response = model.generate(
tokenizer(prompt, return_tensors="pt").input_ids,
logits_processor=[processor],
max_new_tokens=32768,
do_sample=True,
temperature=0.7
)
vLLM
from vllm import LLM, SamplingParams
from scripts.nowait_processor import get_nowait_bad_words_ids
llm = LLM(model="Qwen/QwQ-32B")
bad_words_ids = get_nowait_bad_words_ids(llm.get_tokenizer())
sampling_params = SamplingParams(
max_tokens=32768,
bad_words_ids=bad_words_ids
)
Expected Results
| Task Type | Original Tokens | NOWAIT Tokens | Reduction | |-----------|-----------------|---------------|-----------| | Math (AIME) | 15,000 | 10,500 | 30% | | Visual QA (MMMU) | 2,900 | 1,450 | 50% | | Video QA (MMVU) | 1,700 | 1,250 | 27% |
Limitations
- Less effective on very simple problems where CoT overhead is already minimal
- Distilled models may suffer accuracy loss on challenging tasks
- Some domains may require model-specific keyword tuning
References
- Paper: arXiv:2506.08343v2
- Complete keyword list:
references/keywords.md - Implementation:
scripts/nowait_processor.py
Related Skills
claude-mem
117.6kPersistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
Understand-Anything
92.8kGraphs that teach > graphs that impress. Turn any code into an interactive knowledge graph you can explore, search, and ask questions about. Works with Claude Code, Codex, Cursor, Copilot, Gemini CLI, and more.
headroom
79.8kCompress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
CowAgent
48.0kOpen-source personal AI assistant & Agent Harness. Plans tasks, runs tools and skills, self-evolves with memory and knowledge. Multi-agent, multi-model, multi-channel. Lightweight, extensible, one-line install.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.