List your company

Function Calling on BFCL V3

79.3Overall AccuracyD-CORE-14B
37.64848.461559.27570.0885Mar 10, 2025May 15, 2025Jul 21, 2025Sep 26, 2025Dec 2, 2025Feb 7, 2026Apr 15, 2026

Evaluation results

MethodLinks
D-CORE-14BModel Category=Custom-Trained Model, Backbone=Qwen3-14B, Parameters=14B2026.0279.3——82.967.489.272.2————87.2—————————————
HardGen-4B-RLAccess=Open-source, Parameter Counts=4B, Training=Reinforcement Learning2026.0179.14——————68.564.550.569———90.7383.5587.1463.13———————
Claude-Opus-4-5-20251101Access=Closed-source, Parameter Counts=-2026.0178.92——————81645870.5———88.5879.7984.1968.38———————
GLM-4.6Access=Open-source, Parameter Counts=355B2026.0178.82——————74.5686366.5———87.5680.984.2368———————
xLAM2-70BModel Category=Open-Source Model, Parameters=70B2026.0278.4——72.97578.966.7————88.4—————————————
Gemini-3-Pro-PreviewAccess=Closed-source, Parameter Counts=-2026.0178.17——————64.56054.564———90.6583.1286.8960.75———————
D-CORE-8BModel Category=Custom-Trained Model, Backbone=Qwen3-8B, Parameters=8B2026.0277.7——82.263.888.277.8————86.4—————————————
Claude-Sonnet-4-5-20250929Access=Closed-source, Parameter Counts=-2026.0177.05——————696552.559———88.6581.1384.8961.38———————
xLAM2-32BModel Category=Open-Source Model, Parameters=32B2026.0276.4——74.267.176.788.9————89.3—————————————
Grok-4-1-fast-reasoningAccess=Closed-source, Parameter Counts=-2026.0175.2——————70.559.54362.5———88.2778.4683.3758.88———————
WATT-TOOL-70BScale=70B, Protocol=FC2025.0374.3184.0689.3977.7458.7594.4476.3267.557.548.561.5——————————————
HardGen-4B-SFTAccess=Open-source, Parameter Counts=4B, Training=Supervised Fine-Tuning2026.0173.7——————55.249.141.452.9———89.0382.4285.7349.65———————
Claude-Haiku-4-5-20251001Access=Closed-source, Parameter Counts=-2026.0172.93——————63.542.552.556———86.578.6882.5953.63———————
Qwen3-32BAccess=Open-source, Parameter Counts=32B2026.0172.88——————5652.54043———88.7782.0185.3947.88———————
GPT-4o-2024-11-20Protocol=PROMPT2025.0372.0888.189.3879.8347.6283.3383.76594135.555——————————————
xLAM2-8BModel Category=Open-Source Model, Parameters=8B2026.0272——66.769.164.377.8————84.4—————————————
GPT-4oModel Category=Proprietary Model2026.0271.7——78.85081.383.3————86.8—————————————
Qwen3-235B-A22B-InstructAccess=Open-source, Parameter Counts=235B2026.0171.21——————5442.531.550.5———90.3378.6884.5144.63———————
Grok-4-1-fast-non-reasoningAccess=Closed-source, Parameter Counts=-2026.0170.94——————5839.537.552———88.1377.9483.0446.75———————
Nanbeige4-3B-Thinking-2511Access=Open-source, Parameter Counts=3B2026.0170.71——————58.5544547———81.5879.4280.551.12———————
Kimi-K2-InstructAccess=Open-source, Parameter Counts=1043B2026.0170.3——————624144.555———81.678.6880.1450.63———————
AWMModel Size=14B2026.0270.18—81.4677.251.8878.37———————————————————
GPT-4o-2024-11-20Protocol=FC2025.0369.5887.4289.279.654183.3383.1562.5637.558——————————————
Qwen3-32BModel Category=Open-Source Model, Parameters=32B2026.0269.2——77.343.175.872.2————88.9—————————————
DeepSeek-V3.2-ExpAccess=Open-source, Parameter Counts=671B2026.0168.81——————55492748.5———85.5276.0280.7744.88———————
ToolRL-Qwen3-14BModel Category=Custom-Trained Model, Backbone=Qwen3-14B2026.0268.5——81.535.881.972.2————89.4—————————————
xLAM-2-3b-fc-rAccess=Open-source, Parameter Counts=3B2026.0168.09——————71.55957.545.5———82.9662.9272.9458.38———————
MAGNET-14B-mDPOScale=14B, Strategy=mDPO2025.0368.0190.1389.7579.1437.8888.8984.78523635.528——————————————
GPT-4-TURBO-2024-04-092025.0367.8884.7385.2180.538.1272.2283.815413.535.549.5——————————————
o1Model Category=Proprietary Model2026.0267.8——80.636.187.872.2————85.7—————————————
Watt-ToolParameters=8B2026.0267.79————————————————————————
TeacherParameters=8B2026.0267.74————————————————————————
SimulatorModel Size=14B2026.0267.68—77.2373.4352.3876.91———————————————————
ToolACE-2-8BAccess=Open-source, Parameter Counts=8B2026.0167.63——————492830.546———87.177.4282.2638.38———————
WATT-TOOL-8B*Scale=8B, Protocol=FC, reproduced=true2025.0367.3386.4487.7376.2338.2577.7882.8946402740——————————————
MAGNET-14B-SFTScale=14B, Strategy=SFT2025.0366.8390.0288.277.9233.3872.2282.5947323222.5——————————————
o1-2024-12-17Protocol=PROMPT2025.0366.7378.9282.778.1428.2561.1189.6240.5534.533——————————————
Qwen3Parameters=8B2026.0266.34————————————————————————
Qwen3-8BModel Category=Open-Source Model, Parameters=8B2026.0266.3——78.53379.177.8————88.8—————————————
AWMModel Size=8B2026.0265.94—80.4472.394570.8———————————————————
Qwen3-14BModel Category=Open-Source Model, Parameters=14B2026.0265.9——8036.68172.2————88.2—————————————
ToolRL-Qwen3-8BModel Category=Custom-Trained Model, Backbone=Qwen3-8B2026.0265.9——82.426.885.977.8————88.9—————————————
GPT-4o-2024-11-20Access=Closed-source, Parameter Counts=-2026.0165.64——————55.534.52951———83.8870.5477.2142.5———————
ToolACE-MTAccess=Open-source, Parameter Counts=8B2026.0165.57——————57.531.53438———84.9471.5278.2340.25———————
STARParameters=4B2026.0265.24————————————————————————
Gemini-2.5-FlashAccess=Closed-source, Parameter Counts=-2026.0165.2——————41.5363235.5———84.9674.3979.6836.25———————
MAGNET-7B-mDPOScale=7B, Strategy=mDPO2025.0364.6489.489.2777.9227.7583.3378.5139242622——————————————
Qwen3-30B-A3B-ThinkingAccess=Open-source, Parameter Counts=30B2026.0164.57——————43.510.52541———85.7777.9481.8630———————
AWMModel Size=4B2026.0264.5—78.676.9139.3873.7———————————————————
DeepSeek-R1Model Category=Proprietary Model2026.0263.8——77.338.973.377.8————75.5—————————————
Qwen3Parameters=4B2026.0263.39————————————————————————
MAGNET-7B-SFTScale=7B, Strategy=SFT2025.0362.7388.685.7374.1926.566.6778.6735.52427.519——————————————
ToolRL+DiPOBackbone=Qwen2.5-7B-Instruct2026.0462.51——76.8324.587.569.57————86.21—————————————
Hammer2.1Parameters=7B2026.0262.25————————————————————————
Gemini-1.5-Pro-002Protocol=Prompt2025.0362.1988.5891.2776.7220.7572.2278.152319.517.523——————————————
Qwen3-4B (Base Model)Access=Open-source, Parameter Counts=4B2026.0162.13——————26.52115.525.5———87.8876.3982.1422.13———————
Hammer2.1-7bScale=7B, Protocol=FC2025.0361.8388.6585.4875.1123.582.3578.5935.525.51914——————————————
Gemini-2.0-Flash-ExpProtocol=Prompt2025.0361.7489.9679.8982.0117.8877.7886.442831921.5——————————————
BaseModel Size=14B2026.0261.25—69.4867.284777.94———————————————————
ToolRL+DAPOBackbone=Qwen2.5-7B-Instruct2026.0461.06——76.3119.7587.567.25————87.1—————————————
Llama-4-MaverickAccess=Open-source, Parameter Counts=400B2026.0160.85——————27221418———88.6573.6581.1520.25———————
GPT-5.2-2025-12-11Access=Closed-source, Parameter Counts=-2026.0160.12——————36.51827.530.5———81.8570.3976.1228.13———————
Hammer2.1Parameters=3B2026.0259.56————————————————————————
Claude-3.7-SonnetModel Category=Proprietary Model2026.0258.6——78.448.481.472.2————41.3—————————————
ToolRL+GRPOBackbone=Qwen2.5-7B-Instruct2026.0458.38——74.918.1283.3376.68————86.17—————————————
Functionary-Small-v3.1Scale=7B, Protocol=FC2025.0356.4986.7587.1273.7510.1277.7870.89182.5146——————————————
STARParameters=1.7B2026.0256.05————————————————————————
SimulatorModel Size=4B2026.0255.52—66.162.6937.7575.48———————————————————
ToolRL+DiPOBackbone=Qwen2.5-3B-Instruct2026.0455.03——73.068.6210054.16————83.42—————————————
BaseModel Size=4B2026.0254.92—61.4463.9539.3873.93———————————————————
Qwen3Parameters=1.7B2026.0254.7————————————————————————
EnvScalerModel Size=4B2026.0254.06—60.3164.2537.6385.25———————————————————
BaseModel Size=8B2026.0253.83—59.5858.0343.8876.42———————————————————
ToolRL+DAPOBackbone=Qwen2.5-3B-Instruct2026.0453.21——69.43881.2557.6————82.19—————————————
Qwen2.5-Coder-7B-InstructScale=7B2025.0353.1386.8382.2766.998.2588.8965.3911.56.55.55.5——————————————
ToolRL+GRPOBackbone=Qwen2.5-3B-Instruct2026.0452.98——73.783.7510056.44————81.58—————————————
SimulatorModel Size=8B2026.0252.53—56.3559.3641.8874.06———————————————————
Qwen2.5-Coder-14B-InstructScale=14B2025.0351.8890.9487.865.35.3810044.587.5743——————————————
STARParameters=0.6B2026.0251.7————————————————————————
CKD+Sim-RLBackbone=Qwen3-0.6B, Framework=STAR2026.0251.7——68.197——————78.65—————————————
PPO, Cold StartBackbone=Qwen2.5-3B-Instruct2026.0451.15——67.784.8810018.09————82.42—————————————
SFT+Sim-RLBackbone=Qwen3-0.6B, Note=Sim-RL on SFT-think2026.0250.41——66.996.13——————76.27—————————————
CKDBackbone=Qwen3-0.6B2026.0249.84——66.155.62——————75.92—————————————
Llama3.1Parameters=8B2026.0249.57————————————————————————
FKLBackbone=Qwen3-0.6B2026.0249.51——65.935.12——————76.44—————————————
Sim-RLBackbone=Qwen3-0.6B2026.0249.35——67.393.25——————75.21—————————————
LUFFYBackbone=Qwen3-0.6B2026.0249.23——64.595.48——————76.75—————————————
SFT-thinkBackbone=Qwen3-0.6B2026.0247.59——64.464.5——————71.54—————————————
ToolRLBackbone=Qwen3-0.6B2026.0247.35——66.556.75——————64.81—————————————
Qwen3Parameters=0.6B2026.0247.33————————————————————————
Base-modelBackbone=Qwen3-0.6B2026.0247.33——65.661.88——————71.81—————————————
GKDBackbone=Qwen3-0.6B2026.0247.32——67.613.25——————67.62—————————————
PPO, Cold StartBackbone=Qwen2.5-7B-Instruct2026.0446.68——63.170.3888.8952.92————79.33—————————————
SFT400+GRPOBackbone=Qwen2.5-3B-Instruct2026.0446.42——64.151.7594.4458.63————76.21—————————————
Llama3.2Parameters=3B2026.0245.86————————————————————————
SFT400+PPOBackbone=Qwen2.5-3B-Instruct2026.0445.8——58.765.1210048.4————78.29—————————————
SFTBackbone=Qwen3-0.6B2026.0244.58——62.151.62——————66.29—————————————
SFT400+PPOBackbone=Qwen2.5-7B-Instruct2026.0442.02——51.840.2510029.66————83.9—————————————
Base modelBackbone=Qwen2.5-7B-Instruct2026.0441.97——53.514.2576.4762.66————66.02—————————————
SFT400+GRPOBackbone=Qwen2.5-7B-Instruct2026.0439.25——46.510.2510014.19————80.69—————————————