All issues

[BFCL] Strict regex-based evaluator gives false negatives even on semantically accurate model responses

#1355

· Viranshu-30 opened

on Aug 7, 2026

[BFCL] simple_python_363: release-commit answer key expects `find_closest`, but the item only presents `restaurant_search.find_closest`

#1354

· dipak-bhujbal opened

on Aug 5, 2026

BFCL: deliberately skipped categories render as 0.00% in score CSVs and dilute Overall Acc

#1353

· Blackwellboy opened

on Aug 5, 2026

[BFCL] Clarify default and ablation behavior of synthetic request failures in BFCL V4 Web Search

#1352

· tim2603 opened

on Aug 4, 2026

EvalScope integration: one-command evaluation support for BFCL-v3 and BFCL-v4

#1349

· Yunnglin opened

on Jul 24, 2026

[BFCL] Question regarding Custom Handler implementation for Qwen2.5-Instruct fine-tuned on Rlla-4k (BFCL-v4)

#1343

· EnlightenedAI opened

on Jun 10, 2026

[BFCL] OpenAICompletionsHandler: stale keep-alive connections cause spurious 400 errors

#1339

· jhughes2112 opened

on May 31, 2026

[BFCL] Eager import of qwen_agent forces audio deps on Claude-only users

#1338

· hidingtonothing opened

on May 28, 2026

[BFCL] DeepSeek v4?

#1333

· SamYuan1990 opened

on May 4, 2026

[BFCL] Function-calling parser returns empty response when tool_calls=[]

#1331

· ankush13r opened

on Apr 27, 2026

[BFCL] "Failed to decode the model response" and "Empty response from the model" errors in BFCL4 evaluations f

#1323

· huynhquangdung1210-hue opened

on Apr 7, 2026

[BFCL] Ground truth rejects valid enum values in live_parallel and live_simple

#1314

· bhavishya-pohani opened

on Mar 18, 2026