research-bigcode-evaluation-harness

A tool for testing code-generation models on standard programming challenges such as HumanEval, MBPP, and MultiPL-E. It runs generated code against benchmark tests and reports pass@k, the chance that at least one of several attempts passes.

In plain words
What is it for?
Use it to benchmark a coding model, run selected evaluation suites, save generated answers, and compare results across models or settings.
Why use it?
It provides a repeatable way to compare how well different models generate working code across languages and task sets.

Skill for Claude CodeCodex

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.

agentmods
npx agentmods add skills/graycodeai/starling/research-bigcode-evaluation-harness
Any agent
npx skills add GrayCodeAI/starling --skill research-bigcode-evaluation-harness
Clone the repo
git clone --depth 1 https://github.com/GrayCodeAI/starling

Made for: Claude Code, Codex.

Per session 50 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,213 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin 92% 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 $0.00050 $0.03213
Opus 5 $0.00025 $0.01606
Sonnet 5 $0.00010 $0.00643
Haiku 4.5 $0.00005 $0.00321

Measured 2d ago against content hash 7d3518053def, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

research-bigcode-evaluation-harness scanned grade A with 0 findings 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 2d 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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

This is a copy

92% identical to evaluating-code-models — 9 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.

categories/ai-ml/research-bigcode-evaluation-harness/SKILL.md · 405 lines

How it starts

The opening of the file, as written. The whole thing — 405 lines — stays where its author put it; the contents beside it link to each section on GitHub.

BigCode Evaluation Harness - Code Model Benchmarking

Quick Start

BigCode Evaluation Harness evaluates code generation models across 15+ benchmarks including HumanEval, MBPP, and MultiPL-E (18 languages).

Installation:

git clone https://github.com/bigcode-project/bigcode-evaluation-harness.git
cd bigcode-evaluation-harness
pip install -e .
accelerate config

Evaluate on HumanEval:

accelerate launch main.py \
  --model bigcode/starcoder2-7b \
  --tasks humaneval \
  --max_length_generation 512 \
  --temperature 0.2 \
  --n_samples 20 \
  --batch_size 10 \
  --allow_code_execution \
  --save_generations

View available tasks:

python -c "from bigcode_eval.tasks import ALL_TASKS; print(ALL_TASKS)"

Common Workflows

Workflow 1: Standard Code Benchmark Evaluation

Evaluate model on core code benchmarks (HumanEval, MBPP, HumanEval+).

Checklist:

Code Benchmark Evaluation:
- [ ] Step 1: Choose benchmark suite
- [ ] Step 2: Configure model and generation
- [ ] Step 3: Run evaluation with code execution
- [ ] Step 4: Analyze pass@k results

Step 1: Choose benchmark suite

Python code generation (most common):

  • HumanEval: 164 handwritten problems, function completion
  • HumanEval+: Same 164 problems with 80× more tests (stricter)
  • MBPP: 500 crowd-sourced problems, entry-level difficulty
  • MBPP+: 399 curated problems with 35× more tests

Multi-language (18 languages):

  • MultiPL-E: HumanEval/MBPP translated to C++, Java, JavaScript, Go, Rust, etc.

Advanced:

  • APPS: 10,000 problems (introductory/interview/competition)
  • DS-1000: 1,000 data science problems across 7 libraries

Step 2: Configure model and generation

# Standard HuggingFace model
accelerate launch main.py \
  --model bigcode/starcoder2-7b \
  --tasks humaneval \
  --max_length_generation 512 \
  --temperature 0.2 \
  --do_sample True \
  --n_samples 200 \
  --batch_size 50 \
  --allow_code_execution

# Quantized model (4-bit)
accelerate launch main.py \
  --model codellama/CodeLlama-34b-hf \
  --tasks humaneval \
  --load_in_4bit \
  --max_length_generation 512 \
  --allow_code_execution

# Custom/private model
accelerate launch main.py \
  --model /path/to/my-code-model \
  --tasks humaneval \
  --trust_remote_code \
  --use_auth_token \
  --allow_code_execution

Read the full file on GitHub · 405 lines

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. 2d ago First seen · 405 lines · 50 tokens per session scan A 7d3518053def

Subscribe to this mod's changes

research-bigcode-evaluation-harness is a skill published in the GitHub repository GrayCodeAI/starling (2 stars, last pushed 2d ago), licensed MIT. It adds 50 tokens to every session and 3,213 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. It is 92% identical to evaluating-code-models, differing in 9 lines, and is treated as a copy.

Related

Other skills, from other repositories

evaluating-code-models

Evaluates code generation models across HumanEval, MBPP, MultiPL-E, and 15+ benchmarks with pass@k metrics. Use when benchmarking code models, comparing coding abilities, testing multi-language support, or measuring code generation quality. Industry standard from BigCode Project used by HuggingFace leaderboards.

Orchestra-Research/AI-Research-SKILLs · 68 tokens

evaluating-code-models

Evaluates code generation models across HumanEval, MBPP, MultiPL-E, and 15+ benchmarks with pass@k metrics. Use when benchmarking code models, comparing coding abilities, testing multi-language support, or measuring code generation quality. Industry standard from BigCode Project used by HuggingFace leaderboards.

ihatesea69/HieuNghi-AI-Skills · 68 tokens

build-teaql-app

Build or change a TeaQL application in Java, Rust, Go, Swift, Python, C#/.NET, or TypeScript, including Kotlin/JVM applications that consume Java-generated libraries. Mandatory order: first draft and save a complete KSML model, then verify the client and evaluate that saved model, repair it through repeated evaluation…

teaql/teaql-agent-kit · 112 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

nemo-evaluator-sdk

Evaluates LLMs across 100+ benchmarks from 18+ harnesses (MMLU, HumanEval, GSM8K, safety, VLM) with multi-backend execution. Use when needing scalable evaluation on local Docker, Slurm HPC, or cloud platforms. NVIDIA's enterprise-grade platform with container-first architecture for reproducible benchmarking.

Orchestra-Research/AI-Research-SKILLs · 76 tokens

swe-af

Autonomous engineering team runtime — one API call spins up coordinated AI agents to scope, build, and ship software.

Agent-Field/SWE-AF · 27 tokens