OpenBench

Provider-agnostic, open-source evaluation infrastructure for language models 🚀

OpenBench provides standardized, reproducible benchmarking for LLMs across 30+ evaluation suites (and growing) spanning knowledge, math, reasoning, coding, science, reading comprehension, health, long-context recall, graph reasoning, and first-class support for your own local evals to preserve privacy. Works with any model provider - Groq, OpenAI, Anthropic, Cohere, Google, AWS Bedrock, Azure, local models via Ollama, Hugging Face, and 30+ other providers.

🚧 Alpha Release

We're building in public! This is an alpha release - expect rapid iteration. The first stable release is coming soon.

🎉 What's New in v0.3.0

• 📡 18 More Model Providers: Added support for AI21, Baseten, Cerebras, Cohere, Crusoe, DeepInfra, Friendli, Hugging Face, Hyperbolic, Lambda, MiniMax, Moonshot, Nebius, Nous, Novita, Parasail, Reka, SambaNova and more
• 🧪 New Benchmarks: DROP (reading comprehension), experimental benchmarks available with --alpha flag
• ⚡ CLI Enhancements: openbench alias, -M/-T flags for model/task args, --debug mode for eval-retry
• 🔧 Developer Tools: GitHub Actions integration, Inspect AI extension support

Features

• 🎯 35+ Benchmarks: MMLU, GPQA, HumanEval, SimpleQA, competition math (AIME, HMMT), SciCode, GraphWalks, and more
• 🔧 Simple CLI: bench list, bench describe, bench eval (also available as openbench)
• 🏗️ Built on inspect-ai: Industry-standard evaluation framework
• 📊 Extensible: Easy to add new benchmarks and metrics
• 🤖 Provider-agnostic: Works with 30+ model providers out of the box
• 🛠️ Local Eval Support: Privatized benchmarks can now be run with bench eval <path>
• 📤 Hugging Face Integration: Push evaluation results directly to Hugging Face datasets

🏃 Speedrun: Evaluate a Model in 60 Seconds

Prerequisite: Install uv

# Create a virtual environment and install OpenBench (30 seconds)
uv venv
source .venv/bin/activate
uv pip install openbench

# Set your API key (any provider!)
export GROQ_API_KEY=your_key  # or OPENAI_API_KEY, ANTHROPIC_API_KEY, etc.

# Run your first eval (30 seconds)
bench eval mmlu --model groq/llama-3.3-70b-versatile --limit 10

# That's it! 🎉 Check results in ./logs/ or view them in an interactive UI:
bench view

openbench.mp4

Using Different Providers

# Groq (blazing fast!)
bench eval gpqa_diamond --model groq/meta-llama/llama-4-maverick-17b-128e-instruct

# OpenAI
bench eval humaneval --model openai/o3-2025-04-16

# Anthropic
bench eval simpleqa --model anthropic/claude-sonnet-4-20250514

# Google
bench eval mmlu --model google/gemini-2.5-pro

# Local models with Ollama
bench eval musr --model ollama/llama3.1:70b

# Hugging Face Inference Providers
bench eval mmlu --model huggingface/gpt-oss-120b:groq

# 30+ providers supported - see full list below

Supported Providers

OpenBench supports 30+ model providers through Inspect AI. Set the appropriate API key environment variable and you're ready to go:

Provider	Environment Variable	Example Model String
AI21 Labs	AI21_API_KEY	ai21/model-name
Anthropic	ANTHROPIC_API_KEY	anthropic/model-name
AWS Bedrock	AWS credentials	bedrock/model-name
Azure	AZURE_OPENAI_API_KEY	azure/<deployment-name>
Baseten	BASETEN_API_KEY	baseten/model-name
Cerebras	CEREBRAS_API_KEY	cerebras/model-name
Cohere	COHERE_API_KEY	cohere/model-name
Crusoe	CRUSOE_API_KEY	crusoe/model-name
DeepInfra	DEEPINFRA_API_KEY	deepinfra/model-name
Friendli	FRIENDLI_TOKEN	friendli/model-name
Google	GOOGLE_API_KEY	google/model-name
Groq	GROQ_API_KEY	groq/model-name
Hugging Face	HF_TOKEN	huggingface/model-name
Hyperbolic	HYPERBOLIC_API_KEY	hyperbolic/model-name
Lambda	LAMBDA_API_KEY	lambda/model-name
MiniMax	MINIMAX_API_KEY	minimax/model-name
Mistral	MISTRAL_API_KEY	mistral/model-name
Moonshot	MOONSHOT_API_KEY	moonshot/model-name
Nebius	NEBIUS_API_KEY	nebius/model-name
Nous Research	NOUS_API_KEY	nous/model-name
Novita AI	NOVITA_API_KEY	novita/model-name
Ollama	None (local)	ollama/model-name
OpenAI	OPENAI_API_KEY	openai/model-name
OpenRouter	OPENROUTER_API_KEY	openrouter/model-name
Parasail	PARASAIL_API_KEY	parasail/model-name
Perplexity	PERPLEXITY_API_KEY	perplexity/model-name
Reka	REKA_API_KEY	reka/model-name
SambaNova	SAMBANOVA_API_KEY	sambanova/model-name
Together AI	TOGETHER_API_KEY	together/model-name
Vercel AI Gateway	AI_GATEWAY_API_KEY	vercel/creator-name/model-name
vLLM	None (local)	vllm/model-name

Available Benchmarks

Here are the currently available benchmarks. For an up-to-date list use bench list.

Category	Benchmarks
Knowledge	MMLU (57 subjects), GPQA (graduate-level), SuperGPQA (285 disciplines), OpenBookQA, HLE (Humanity's Last Exam - 2,500 questions from 1,000+ experts), HLE_text (text-only version)
Coding	HumanEval (164 problems)
Math	AIME 2023-2025, HMMT Feb 2023-2025, BRUMO 2025, MATH (competition-level problems), MATH-500 (challenging subset), MGSM (multilingual grade school math), MGSM_en (English), MGSM_latin (5 languages), MGSM_non_latin (6 languages)
Reasoning	SimpleQA (factuality), MuSR (multi-step reasoning), DROP (discrete reasoning over paragraphs)
Long Context	OpenAI MRCR (multiple needle retrieval), OpenAI MRCR_2n (2 needle), OpenAI MRCR_4 (4 needle), OpenAI MRCR_8n (8 needle)
Healthcare	HealthBench (open-ended healthcare eval), HealthBench_hard (challenging variant), HealthBench_consensus (consensus variant)
Cybersecurity	CTI-Bench (complete cyber threat intelligence suite), CTI-Bench ATE (MITRE ATT&CK technique extraction), CTI-Bench MCQ (knowledge questions on CTI standards and best practices), CTI-Bench RCM (CVE to CWE vulnerability mapping), CTI-Bench VSP (CVSS score calculation)

Configuration

# Set your API keys
export GROQ_API_KEY=your_key
export HF_TOKEN=your_key
export OPENAI_API_KEY=your_key  # Optional

# Set default model
export BENCH_MODEL=groq/llama-3.1-70b

Commands and Options

For a complete list of all commands and options, run: bench --help

Command	Description
bench or openbench	Show main menu with available commands
bench list	List available evaluations, models, and flags
bench eval <benchmark>	Run benchmark evaluation on a model
bench eval-retry	Retry a failed evaluation
bench view	View logs from previous benchmark runs
bench eval <path>	Run your local/private evals built with Inspect AI

Key eval Command Options

Option	Environment Variable	Default	Description
-M <args>	None	None	Pass model-specific arguments (e.g., -M reasoning_effort=high)
-T <args>	None	None	Pass task-specific arguments to the benchmark
--model	BENCH_MODEL	None (required)	Model(s) to evaluate
--epochs	BENCH_EPOCHS	1	Number of epochs to run each evaluation
--max-connections	BENCH_MAX_CONNECTIONS	10	Maximum parallel requests to model
--temperature	BENCH_TEMPERATURE	0.6	Model temperature
--top-p	BENCH_TOP_P	1.0	Model top-p
--max-tokens	BENCH_MAX_TOKENS	None	Maximum tokens for model response
--seed	BENCH_SEED	None	Seed for deterministic generation
--limit	BENCH_LIMIT	None	Limit evaluated samples (number or start,end)
--logfile	BENCH_OUTPUT	None	Output file for results
--sandbox	BENCH_SANDBOX	None	Environment to run evaluation (local/docker)
--timeout	BENCH_TIMEOUT	10000	Timeout for each API request (seconds)
--display	BENCH_DISPLAY	None	Display type (full/conversation/rich/plain/none)
--reasoning-effort	BENCH_REASONING_EFFORT	None	Reasoning effort level (low/medium/high)
--json	None	False	Output results in JSON format
--hub-repo	BENCH_HUB_REPO	None	Push results to a Hugging Face Hub dataset

Building Your Own Evals

OpenBench is built on Inspect AI. To create custom evaluations, check out their excellent documentation. Once you do build your own private evaluations with Inspect AI that you don't want to open-source, you can point OpenBench at them with bench eval <path> to run!

Exporting Logs to Hugging Face

OpenBench can export logs to a Hugging Face Hub dataset. This is useful if you want to share your results with the community or use them for further analysis.

export HF_TOKEN=<your-huggingface-token>

bench eval mmlu --model groq/llama-3.3-70b-versatile --limit 10 --hub-repo <your-username>/openbench-logs 

This will export the logs to a Hugging Face Hub dataset with the name openbench-logs.

FAQ

How does OpenBench differ from Inspect AI?

OpenBench provides:

• Reference implementations of 20+ major benchmarks with consistent interfaces
• Shared utilities for common patterns (math scoring, multi-language support, etc.)
• Curated scorers that work across different eval types
• CLI tooling optimized for running standardized benchmarks

Think of it as a benchmark library built on Inspect's excellent foundation.

Why not just use Inspect AI, lm-evaluation-harness, or lighteval?

Different tools for different needs! OpenBench focuses on:

• Shared components: Common scorers, solvers, and datasets across benchmarks reduce code duplication
• Clean implementations: Each eval is written for readability and reliability
• Developer experience: Simple CLI, consistent patterns, easy to extend

We built OpenBench because we needed evaluation code that was easy to understand, modify, and trust. It's a curated set of benchmarks built on Inspect AI's excellent foundation.

How can I run bench outside of the uv environment?

If you want bench to be available outside of uv, you can run the following command:

uv run pip install -e .

I'm running into an issue when downloading a dataset from HuggingFace - how do I fix it?

Some evaluations may require logging into HuggingFace to download the dataset. If bench prompts you to do so, or throws "gated" errors, defining the environment variable

HF_TOKEN="<HUGGINGFACE_TOKEN>"

should fix the issue. The full HuggingFace documentation can be found on the HuggingFace docs on Authentication.

Development

For development work, you'll need to clone the repository:

# Clone the repo
git clone https://github.com/groq/openbench.git
cd openbench

# Setup with UV
uv venv && uv sync --dev
source .venv/bin/activate

# CRITICAL: Install pre-commit hooks (CI will fail without this!)
pre-commit install

# Run tests
pytest

⚠️ IMPORTANT: You MUST run pre-commit install after setup or CI will fail on your PRs!

Contributing

We welcome contributions! Please see our Contributing Guide for detailed instructions on:

• Setting up the development environment
• Adding new benchmarks and model providers
• Code style and testing requirements
• Submitting issues and pull requests

Quick links:

• Report a bug
• Request a feature

Reproducibility Statement

As the authors of OpenBench, we strive to implement this tool's evaluations as faithfully as possible with respect to the original benchmarks themselves.

However, it is expected that developers may observe numerical discrepancies between OpenBench's scores and the reported scores from other sources.

These numerical differences can be attributed to many reasons, including (but not limited to) minor variations in the model prompts, different model quantization or inference approaches, and repurposing benchmarks to be compatible with the packages used to develop OpenBench.

As a result, OpenBench results are meant to be compared with OpenBench results, not as a universal one-to-one comparison with every external result. For meaningful comparisons, ensure you are using the same version of OpenBench.

We encourage developers to identify areas of improvement and we welcome open source contributions to OpenBench.

Acknowledgments

This project would not be possible without:

• Inspect AI - The incredible evaluation framework that powers OpenBench
• EleutherAI's lm-evaluation-harness - Pioneering work in standardized LLM evaluation
• Hugging Face's lighteval - Excellent evaluation infrastructure

Citation

@software{openbench,
  title = {OpenBench: Provider-agnostic, open-source evaluation infrastructure for language models},
  author = {Sah, Aarush},
  year = {2025},
  url = {https://openbench.dev}
}

License

MIT

---

Built with ❤️ by Aarush Sah and the Groq team