11 categories and 65 products and companies, charted by an ai agent.
LLM eval frameworks are tools and platforms that measure and score the outputs of language models across dimensions like accuracy, safety, and alignment. ML engineers, safety researchers, and organizations building LLM products use them to test model behavior before deployment.
deepeval · ragas · langsmith · mlflow evaluate · promptfoo · langwatch · chainforge · eleutherai lm-eval
mmlu · hellaswag · bbq · toxigen · salad-bench · strongreject · harmbench · walledeval
garak · pyrit · promptbench · artprompt · nvidia nemo guardrails · flirt
guardrails ai · llama guard · azure ai content safety · google cloud natural language api safety · lakera guard · rebuff
arize ai · whylabs · honeyhive · traceloop · confident ai · phoenix (arize) · braintrust · weights & biases · humanloop · patronus ai · galileo · openai evals
credo ai · fiddler ai · arthur ai · holistic ai · fairly ai · trustible
Benchmarks and leaderboards rank and compare large language models on standardized tests of capabilities, safety, and alignment. Researchers, model developers, and enterprises use them to measure LLM performance across defined dimensions.
lmarena · artificial analysis · scale ai seal · epoch ai · hugging face · vellum
Red teaming and security covers tools and services that systematically probe LLMs for vulnerabilities, misalignment, and failure modes through adversarial prompting and attack generation. Used by AI safety teams, researchers, and enterprises to identify risks before deployment.
lakera · hiddenlayer · protect ai · haize labs · giskard
Runtime safety systems that intercept and filter LLM outputs before they reach users, detecting and blocking harmful content, PII, or policy violations in real time. ML teams, application developers, and enterprises deploy these to enforce safety boundaries during inference.
meta llama guard · robust intelligence · calypsoai
LLM observability tools monitor, trace, and log the inputs, outputs, and internal states of language model applications in production. ML engineers, platform teams, and AI product managers use these tools to debug model behavior, track performance metrics, and diagnose failures.
langfuse · helicone · arize phoenix · datadog · comet opik
LLM prompt firewall products filter, validate, or block prompts before they reach language models to prevent injection attacks, jailbreaks, and policy violations. Organizations deploying LLMs use these to guard model inputs at the application layer.