Search Results for "benchmark testing"

...These environments require agents to interpret instructions, take actions, and adapt their strategies based on feedback from the environment. AgentBench also includes an evaluation framework that measures success rates, rewards, and task completion performance across different agent implementations. By testing models across diverse scenarios, the benchmark highlights strengths and weaknesses in reasoning, long-term planning, and tool usage.

AICGSecEval is an open-source benchmark framework designed to evaluate the security of code generated by artificial intelligence systems. The project was developed to address concerns that AI-assisted programming tools may produce insecure code containing vulnerabilities such as injection flaws or unsafe logic. The framework constructs evaluation tasks based on real-world software repositories and known vulnerability cases derived from CVE records. By simulating realistic development...

Hallucination Leaderboard is an open research project that tracks and compares the tendency of large language models to produce hallucinated or inaccurate information when generating summaries. The project provides a standardized benchmark that evaluates different models using a dedicated hallucination detection system known as the Hallucination Evaluation Model. Each model is tested on document summarization tasks to measure how often generated responses introduce information that is not...

...It provides multiple entry points for developers, including a CLI for running models locally and an HTTP server that exposes an OpenAI-compatible API surface for easy integration with existing clients. The project includes hardware-aware tooling that can benchmark a system and choose sensible quantization and device-mapping strategies, helping users get strong performance without manual tuning. It also supports serving multiple models from the same server process, enabling routing or quick switching between models depending on workload needs. For user-facing testing, mistral.rs can provide a built-in web UI, and it also offers a dedicated lightweight web chat interface that supports richer interaction patterns.

MiniMax-M2.5 is a state-of-the-art foundation model extensively trained with reinforcement learning across hundreds of thousands of real-world environments. It delivers leading performance in coding, agentic tool use, search, and complex office workflows, achieving top benchmark scores such as 80.2% on SWE-Bench Verified and 76.3% on BrowseComp. Designed to reason efficiently and decompose tasks like an experienced architect, M2.5 plans features, structure, and system design before generating code. The model supports full-stack development across web, mobile, and desktop platforms, covering the entire lifecycle from system design to testing and code review. ...

Safety-Prompts is an open-source repository that provides a curated collection of prompts designed to evaluate and improve the safety behavior of large language models. The project focuses primarily on safety testing scenarios relevant to Chinese language models, though the concepts can be applied to other languages and systems. The prompts are structured to test whether models generate outputs that align with human values and safety guidelines when faced with potentially harmful or sensitive requests. Researchers and developers use the dataset to benchmark how well models avoid unsafe responses and follow alignment constraints. ...

The grade-school-math repository (sometimes called GSM8K) is a curated dataset of 8,500 high-quality grade school math word problems intended for evaluating mathematical reasoning capabilities of language models. It is structured into 7,500 training problems and 1,000 test problems. These aren’t trivial exercises — many require multi-step reasoning, combining arithmetic operations, and handling intermediate steps (e.g. “If she sold half as many in May… how many in total?”). The problems are...