evaluating-llms-harness

evaluating-llms-harness is a skill for Claude Code, Codex from OpenRaiser/NanoResearch. It costs 85 tokens per session (3,474 once invoked), scanned A, a copy of evaluating-llms-harness, MIT.

A tool for measuring how well language models perform on standard academic tests, including tests for knowledge, reasoning, truthfulness, and code generation. It supports models hosted locally with Hugging Face or vLLM as well as models accessed through APIs.

In plain words
What is it for?
Use it to run benchmarks such as MMLU, GSM8K, HumanEval, HellaSwag, TruthfulQA, and ARC, then review the results.
Why use it?
It gives you consistent test results for comparing models or checking whether a model improves, rather than relying on informal examples.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: positional $N argument.

Good fit Use it to run benchmarks such as MMLU, GSM8K, HumanEval, HellaSwag, TruthfulQA, and ARC, then review the results.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/openraiser/nanoresearch/lm-evaluation-harness
About the project

NanoResearch is an autonomous AI research system that turns research ideas into executable experiments and LaTeX papers supported by results from real training runs. It is for researchers validating prototypes, running GPU experiments, generating benchmarks, analyzing logs, and preparing paper drafts. The catalogue add-ons support its research pipeline and agent workflows.

OpenRaiser/NanoResearch · 1,365 stars · on GitHub

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add OpenRaiser/NanoResearch --skill lm-evaluation-harness
Clone the repo
git clone --depth 1 https://github.com/OpenRaiser/NanoResearch

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for evaluating-llms-harness

README.md
[![agentmods](https://agentmods.dev/badge/skills/openraiser/nanoresearch/lm-evaluation-harness/github.svg)](https://agentmods.dev/skills/openraiser/nanoresearch/lm-evaluation-harness)
Your own site
<a href="https://agentmods.dev/skills/openraiser/nanoresearch/lm-evaluation-harness"><img src="https://agentmods.dev/badge/skills/openraiser/nanoresearch/lm-evaluation-harness/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for evaluating-llms-harness

Your own site · 80×15
<a href="https://agentmods.dev/skills/openraiser/nanoresearch/lm-evaluation-harness"><img src="https://agentmods.dev/badge/skills/openraiser/nanoresearch/lm-evaluation-harness.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 85 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,474 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.
Origin 89% copy Near-identical to another mod in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00085 $0.03474
Opus 5 $0.00043 $0.01737
Sonnet 5 $0.00017 $0.00695
Haiku 4.5 $0.00009 $0.00347

Measured 10d ago against content hash 891b77bc0911, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

evaluating-llms-harness scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

os.system(f"./eval_checkpoint.sh checkpoints step-{step}")
Origin

This is a copy

89% identical to evaluating-llms-harness — 28 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

skills/vendor-ai-research/lm-evaluation-harness/SKILL.md · 491 lines

How it starts

The opening of the file, as written. The whole thing — 491 lines — stays where its author put it; the contents beside it link to each section on GitHub.

lm-evaluation-harness - LLM Benchmarking

Quick start

lm-evaluation-harness evaluates LLMs across 60+ academic benchmarks using standardized prompts and metrics.

Installation:

pip install lm-eval

Evaluate any HuggingFace model:

lm_eval --model hf \
  --model_args pretrained=meta-llama/Llama-2-7b-hf \
  --tasks mmlu,gsm8k,hellaswag \
  --device cuda:0 \
  --batch_size 8

View available tasks:

lm_eval --tasks list

Common workflows

Workflow 1: Standard benchmark evaluation

Evaluate model on core benchmarks (MMLU, GSM8K, HumanEval).

Copy this checklist:

Benchmark Evaluation:
- [ ] Step 1: Choose benchmark suite
- [ ] Step 2: Configure model
- [ ] Step 3: Run evaluation
- [ ] Step 4: Analyze results

Step 1: Choose benchmark suite

Core reasoning benchmarks:

  • MMLU (Massive Multitask Language Understanding) - 57 subjects, multiple choice
  • GSM8K - Grade school math word problems
  • HellaSwag - Common sense reasoning
  • TruthfulQA - Truthfulness and factuality
  • ARC (AI2 Reasoning Challenge) - Science questions

Code benchmarks:

  • HumanEval - Python code generation (164 problems)
  • MBPP (Mostly Basic Python Problems) - Python coding

Standard suite (recommended for model releases):

--tasks mmlu,gsm8k,hellaswag,truthfulqa,arc_challenge

Step 2: Configure model

HuggingFace model:

lm_eval --model hf \
  --model_args pretrained=meta-llama/Llama-2-7b-hf,dtype=bfloat16 \
  --tasks mmlu \
  --device cuda:0 \
  --batch_size auto  # Auto-detect optimal batch size

Quantized model (4-bit/8-bit):

lm_eval --model hf \
  --model_args pretrained=meta-llama/Llama-2-7b-hf,load_in_4bit=True \
  --tasks mmlu \
  --device cuda:0

Custom checkpoint:

lm_eval --model hf \
  --model_args pretrained=/path/to/my-model,tokenizer=/path/to/tokenizer \
  --tasks mmlu \
  --device cuda:0

Step 3: Run evaluation

# Full MMLU evaluation (57 subjects)
lm_eval --model hf \
  --model_args pretrained=meta-llama/Llama-2-7b-hf \
  --tasks mmlu \
  --num_fewshot 5 \  # 5-shot evaluation (standard)
  --batch_size 8 \
  --output_path results/ \
  --log_samples  # Save individual predictions

# Multiple benchmarks at once
lm_eval --model hf \
  --model_args pretrained=meta-llama/Llama-2-7b-hf \
  --tasks mmlu,gsm8k,hellaswag,truthfulqa,arc_challenge \
  --num_fewshot 5 \
  --batch_size 8 \
  --output_path results/llama2-7b-eval.json

Read the full file on GitHub · 491 lines

Files

What ships with it

4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 10d ago First seen · 491 lines · 85 tokens per session scan A 891b77bc0911

Subscribe to this mod's changes

evaluating-llms-harness is a skill published in the GitHub repository OpenRaiser/NanoResearch (1,365 stars, last pushed 15d ago), licensed MIT. It adds 85 tokens to every session and 3,474 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). It is 89% identical to evaluating-llms-harness, differing in 28 lines, and is treated as a copy.

Related

Other skills, from other repositories

evaluating-llms-harness

Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.

sangrokjung/claude-forge · 85 tokens

evaluating-llms-harness

Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.

davila7/claude-code-templates · 85 tokens

evaluating-llms-harness

Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.

OpenLAIR/dr-claw · 85 tokens

evaluating-llms-harness

Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.

foryourhealth111-pixel/Vibe-Skills · 85 tokens

evaluating-llms-harness

Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.

Orchestra-Research/AI-Research-SKILLs · 85 tokens

evaluating-llms-harness

Evaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.

ihatesea69/HieuNghi-AI-Skills · 85 tokens