Search
Everything in the catalog
Showing 1–13 of 13
- 0.762score
τ²-bench (tau-bench)
Sierra
Simulated customer-service benchmark where an agent follows domain policies and uses tools while a simulated user also acts.
ReferenceAFreshness: A★ 2.3K - 0.701score
MCP-Universe
Salesforce AI Research
Benchmark and framework that tests models and agents on tasks using real-world MCP servers.
ReferenceAFreshness: A★ 602 - 0.690score
OTel-2.0-LLM-31B-IT-QLoRA
farbodtavakkoli
Open-weights 31.27B-parameter text model from farbodtavakkoli on Hugging Face (apache-2.0).
ModelAFreshness: AOpen weights - 0.686score
MiniCPM5-2B
openbmb
Open-weights 2.52B-parameter text model from openbmb on Hugging Face (apache-2.0).
ModelAFreshness: AOpen weights - 0.686score
OTel-2.0-LLM-31B-IT
farbodtavakkoli
Open-weights 31.27B-parameter text model from farbodtavakkoli on Hugging Face (apache-2.0).
ModelAFreshness: AOpen weights - 0.683score
Berkeley Function Calling Leaderboard (BFCL)
Gorilla project, UC Berkeley
Leaderboard that tests how accurately models call functions and tools, from single calls to multi-turn agentic use.
ReferenceDFreshness: D★ 13.1K - 0.681score
needle3
Cactus-Compute
Open-weights text model from Cactus-Compute on Hugging Face (apache-2.0).
ModelAFreshness: AOpen weights - 0.641score
MiniCPM5-1B
openbmb
Open-weights 1.08B-parameter text model from openbmb on Hugging Face (apache-2.0).
ModelBFreshness: BOpen weights - 0.641score
Qwen3.8-4B-Distill
empero-ai
Open-weights 4.66B-parameter text model from empero-ai on Hugging Face (apache-2.0).
ModelBFreshness: BOpen weights - 0.581score
MCPMark
EVAL SYS, with LobeHub and NUS TRAIL
Stress test for realistic MCP use across Notion, GitHub, filesystem, PostgreSQL and Playwright servers.
ReferenceCFreshness: C★ 467 - 0.532score
MiniCPM5-2B-DSpark
openbmb
Open-weights 0.32B-parameter text model from openbmb on Hugging Face (apache-2.0).
ModelBFreshness: BOpen weights - 0.462score
dolphin-2.9.1-yi-1.5-34b
dphn
Open-weights 34.39B-parameter text model from dphn on Hugging Face (apache-2.0).
ModelFFreshness: FOpen weights - 0.432score
Hermes-4.3-36B
NousResearch
Open-weights 36.15B-parameter text model from NousResearch on Hugging Face (apache-2.0).
ModelDFreshness: DOpen weights