All pull requests

Pull requests

• Decode unary and binary operators without eval

  #1374·shaurya416 opened Sep 24, 2026·

  ·
  1

• [BFCL] Fix ground truth that contradicts the user or the schema

  #1372·djstrong opened Sep 23, 2026·

  ·

• [BFCL] Accept ints inside float arrays during AST checking

  #1371·djstrong opened Sep 23, 2026·

  ·

• [BFCL] Add solidSF Jesse-Autonomous-v2 (FC & Prompt) model handler and configs

  #1369·m0at opened Sep 19, 2026·

  ·

• [BFCL] Add BitAgent and xLAM DPO-prune models

  #1368·Zanaei1997 opened Sep 19, 2026·

  ·

• Add a handler for Qwen3.5's XML tool-call format

  #1366·ajvikram opened Sep 9, 2026·

  ·
  5

• [BFCL] Defer qwen_agent import until handler initialization

  #1365·Keerthirajan58 opened Sep 4, 2026·

  ·
  1

• fix(bfcl): exclude never-evaluated categories from aggregate scores

  #1364·Huang-404-Q opened Sep 2, 2026·

  ·
  1

• Add BlueLM-Flash to leaderboard

  #1359·liukira459-commits opened Aug 20, 2026·

  ·

• Add AgentWeave + Hammer2.1-1.5B routed BFCL handler

  #1358·sauravsingla opened Aug 18, 2026·

  ·

• [BFCL] Add CloudSurf-4B-FC + Gemma-4 prompt-mode handler fix (stray-closer strip, +4.6 stock E4B)

  #1357·worldMachine opened Aug 17, 2026·

  ·
  1

• Handle textual responses when tool_calls is empty

  #1351·IloDan opened Jul 30, 2026·

  ·

• Update readme file to show the perforamnce of evalscope.

  #1350·HelloWorldLTY opened Jul 30, 2026·

  ·

• fix: model finishing with no tool calls gets misreported as a decode error

  #1348·zambalee opened Jul 18, 2026·

  ·

• [BFCL] Fix ZeroDivisionError in memory kv backend when searching an empty memory

  #1347·e0xextazy opened Jul 16, 2026·

  ·

• [BFCL] Use underscore_to_dot config flag for tool name compilation

  #1346·lukeinglis opened Jun 30, 2026·

  ·
  1

• Add BFCL results for roshangrewal/gemma4-e4b-toolcall-v02

  #1344·roshangrewal opened Jun 16, 2026·

  ·

• Add AYA Pipeline results for BFCL leaderboard

  #1342·Trizz-ai opened Jun 3, 2026·

  ·

• [BFCL] Add LiquidAI LFM2.5 models and LFM2 handlers

  #1341·fatday opened Jun 2, 2026·

  ·

• Add Gemma 4 (26B A4B) prompt-mode handler

  #1340·tumbak opened Jun 1, 2026·

  ·

• feat: add LiteLLM handler for benchmarking any LiteLLM-supported model

  #1337·RheagalFire opened May 17, 2026·

  ·
  2

• Add APIBench eval data drift checker

  #1335·mindbomber opened May 7, 2026·

  ·
  1

• feat(apizoo): add Oracle HCM, SCM, and Procurement Cloud REST APIs

  #1334·karthikchundi-commits opened May 4, 2026·

  ·
  2

• Add Prism Coder 7B Handler

  #1332·dcostenco opened Apr 28, 2026·

  ·

• fix: remove invalid enum value from ground truth in live_multiple_190-84-0

  #1330·octo-patch opened Apr 25, 2026·

  ·
  1