speculative-decoding
speculative-decoding is a Local & open models skill for AI agents, published by davila7 in the claude-code-templates repository on GitHub. Skillspace rates speculative-decoding C (74/100) and it passed the safety scan; a judge model has also reviewed it.
Accelerate LLM inference using speculative decoding, Medusa multiple heads, and lookahead decoding techniques. Use when optimizing inference speed (1.5-3.6× speedup), reducing latency for real-time applications, or deploying models with limited compute. Covers draft models, tree-based attention, Jacobi iteration, parallel token generation, and production deployment strategies.
- Grade
- C
- Quality
- 74 / 100
- Safety
- Passed
- GitHub stars
- 30,694
Placed #5 of 6 in the judged category final for Local & open models.
This rating describes the version of the repository we crawled; it may have changed since, so go by its current files.
Our review
- Architecture
- 96 / 100
- Maturity
- Packaged: licensed, with a version or references
- Security scan
- Low-risk patterns
- License
- Not declared. Check the repository before reusing it.
Judge review
Placed #5 of 6 in the judged category final for Local & open models.
- Trigger
82 82 - Instructions
72 72 - Specificity
85 85 - Safety
62 62 - Disclosure
40 40 - Maintenance
60 60
Strong technical content with real code and paper citations, but a dense wall of text with no caution about installing the third-party repos it clones.
What the scanner flagged
- mentions secrets in text
Not checked yet
- Category is a classifier prediction; no model has read this skill yet
- No blind task eval yet
- Scope (portable / vendor / repo-internal) not assessed; only the top of each category is read for scope
- Sibling files (scripts, references) not fetched; security scan covers SKILL.md only
Skillspace indexes and rates this skill; it does not host or endorse it. Run third-party skills at your own risk.