Leaderboard

Tool Calling Rankings

Function calling and API orchestration in agentic workflows: Toolathlon, MCP Atlas and τ²-Bench Retail.

01
Claude Fable 5
98.5
02
Gemini 3.5 Flash
95.3
03
MiMo V2.5 Pro
94.2
Overall700Coding399Math53Reasoning413Agentic55Knowledge412Multimodal382Coding Agents55Scientific Reasoning82Multilingual12Writing61Chat & Assistants61Research85Tool Calling32Long Context82Speed402Value401Open Weights170
32 models
#ModelPodium ScoreToolathlonMCP Atlasτ²-Bench RetailSpeed
01
Claude Fable 5
Anthropic · reasoning
98.5
——98.5%71 t/s
02
95.3
——95.3%267 t/s
03
MiMo V2.5 Pro
Xiaomi · Open
94.2
——94.2%65 t/s
4
91.5
——91.5%80 t/s
5
Kimi K2.7 Code
Moonshot AI · Open
90.1
——90.1%39 t/s
6
MiniMax M3
MiniMax
88.9
——88.9%93 t/s
7
GPT-6 Astra
OpenAI · reasoning
88.0
——88%82 t/s
8
87.1
——87.1%120 t/s
9
86.0
——86%118 t/s
10
85.6
—76.4%94.7%203 t/s
11
Nemotron 3 Ultra 550B
NVIDIA · Open
83.3
——83.3%142 t/s
12
82.4
—69.2%95.6%136 t/s
13
78.8
59.9%82.2%94.4%50 t/s
14
Kimi K3
Moonshot AI · Open · reasoning
78.7
73.2%84.2%—37 t/s
15
77.3
—77.3%—80 t/s
16
GPT-5.5
OpenAI
74.9
55.6%75.3%93.9%60 t/s
17
GLM-5.2
Z.ai · Open
74.7
48.2%76.8%99.1%164 t/s
18
73.8
—62.7%84.8%50 t/s
19
73.2
—73.2%—54 t/s
20
DeepSeek V4 Pro
DeepSeek · Open
72.2
51.8%73.6%91.2%66 t/s
21
71.6
58%—85.1%72 t/s
22
70.5
39.8%74.1%97.7%55 t/s
23
69.7
53.1%—86.3%127 t/s
24
63.8
48.5%79.1%—69 t/s
25
61.3
—61.3%—50 t/s
26
GPT-5.4
OpenAI
60.9
54.6%67.2%—60 t/s
27
DeepSeek V4 Flash
DeepSeek · Open
58.4
47.8%69%—113 t/s
28
GPT-5.2
OpenAI
53.5
46.3%60.6%—90 t/s
29
53.4
53.4%——175 t/s
30
53.4
49.4%57.4%—120 t/s
31
Kimi K2.6
Moonshot AI · Open
50.0
50%——40 t/s
32
Qwen3.5 397B-A17B
Alibaba · Open
38.3
38.3%——71 t/s
Podium Score is a weighted blend of five leaderboards. See Methodology for details.
0 / 4
Compare Now →