Are there any hebrew tool-calling benchmarks?
#1
by amitgalor - opened
In the wave of agentic applications, are there any Hebrew agentic benchmarks or tool calling benchmarks?
Is anyone working on translating benchmarks such as BFCL or Tau2 to Hebrew?
Also, it would be interesting to see nvidia/Llama-3_3-Nemotron-Super-49B-v1_5 in the leaderboard, as it surpassed llama3.3-70B on several reasoning benchmarks like GPQA, Math 500 and BFCL