AI BENCHY Category
Tool Calling Ranking
See which AI models perform best on Tool Calling, which ones stay reliable, and where the biggest gaps appear. Sort by: Metric ↑.
Related Failure Reasons
| Rank | Model | Company | Tool Calling Score | Avg Score | Tests Correct | Response Time (avg) |
|---|---|---|---|---|---|---|
| #39 | gpt-oss-120b medium | OpenAI | 9.0 | 5.1 | 1/1 | 6.91s |
| #1 | Gemini 3 Flash Preview medium | 10.0 | 10.0 | 1/1 | 10.6s | |
| #2 | Gemini 3.1 Pro Preview medium | 10.0 | 9.4 | 1/1 | 23.1s | |
| #3 | GPT-5.3-Codex medium | OpenAI | 10.0 | 8.4 | 1/1 | 6.37s |
| #4 | Qwen3.5 Plus 2026-02-15 medium | Qwen | 10.0 | 8.3 | 1/1 | 7.54s |
| #5 | Gemini 3 Flash Preview low | 10.0 | 8.2 | 1/1 | 4.99s | |
| #6 | Gemini 3 Pro Preview medium | 10.0 | 8.2 | 1/1 | 12.0s | |
| #7 | Qwen3.5-27B medium | Qwen | 10.0 | 8.2 | 1/1 | 7.45s |
| #8 | Gemini 3.1 Flash Lite Preview high | 10.0 | 8.2 | 1/1 | 7.73s | |
| #9 | GPT-5.4 medium | OpenAI | 10.0 | 8.0 | 1/1 | 13.3s |
| #10 | Qwen3.5-122B-A10B medium | Qwen | 10.0 | 7.7 | 1/1 | 4.60s |
| #11 | Claude Sonnet 4.6 medium | Anthropic | 10.0 | 7.7 | 1/1 | 7.48s |
| #12 | Gemini 3.1 Flash Lite Preview medium | 10.0 | 7.5 | 1/1 | 3.80s | |
| #13 | Step 3.5 Flash medium | Stepfun | 10.0 | 7.4 | 1/1 | 11.9s |
| #14 | GLM 5 medium | Z.ai | 10.0 | 7.4 | 1/1 | 15.9s |
| #15 | GPT-5.2 Chat none | OpenAI | 10.0 | 7.4 | 1/1 | 4.68s |
| #16 | Gemini 2.5 Flash medium | 10.0 | 7.4 | 1/1 | 6.20s | |
| #17 | Gemini 3.1 Flash Lite Preview low | 10.0 | 7.3 | 1/1 | 9.54s | |
| #18 | DeepSeek V3.2 medium | DeepSeek | 10.0 | 7.3 | 1/1 | 34.8s |
| #19 | GPT-5.3 Chat none | OpenAI | 10.0 | 7.3 | 1/1 | 8.36s |
| #20 | Gemini 3 Flash Preview none | 10.0 | 7.2 | 1/1 | 3.35s | |
| #21 | MiMo-V2-Flash medium | Xiaomi | 10.0 | 7.2 | 1/1 | 27.8s |
| #22 | Gemini 3.1 Flash Lite Preview none | 10.0 | 7.1 | 1/1 | 3.39s | |
| #23 | Seed-2.0-Mini medium | Bytedance Seed | 10.0 | 6.9 | 1/1 | 88.7s |
| #24 | Qwen3.5-Flash medium | Qwen | 10.0 | 6.9 | 1/1 | 10.3s |
| #25 | Claude Sonnet 4.6 none | Anthropic | 10.0 | 6.8 | 1/1 | 4.11s |
| #26 | Claude Opus 4.6 medium | Anthropic | 10.0 | 6.6 | 1/1 | 9.73s |
| #27 | GPT-5.2 medium | OpenAI | 10.0 | 6.5 | 0/1 | 10.3s |
| #28 | Kimi K2.5 medium | Moonshot AI | 10.0 | 6.4 | 1/1 | 31.7s |
| #29 | Qwen3.5 Plus 2026-02-15 none | Qwen | 10.0 | 6.2 | 1/1 | 3.33s |
| #30 | Grok 4.1 Fast medium | X AI | 10.0 | 6.2 | 0/1 | 27.7s |
| #31 | GLM 5 none | Z.ai | 10.0 | 6.0 | 1/1 | 11.1s |
| #32 | GPT-5 Mini medium | OpenAI | 10.0 | 6.0 | 1/1 | 18.6s |
| #33 | DeepSeek V3.2 none | DeepSeek | 10.0 | 5.5 | 1/1 | 11.8s |
| #34 | GPT-5 Nano medium | OpenAI | 10.0 | 5.5 | 1/1 | 33.3s |
| #35 | Qwen3.5-35B-A3B medium | Qwen | 10.0 | 5.5 | 1/1 | 4.65s |
| #36 | Mercury 2 medium | Inception | 10.0 | 5.3 | 1/1 | 1.89s |
| #37 | Qwen3.5-Flash none | Qwen | 10.0 | 5.2 | 1/1 | 3.67s |
| #38 | Gemini 2.5 Flash none | 10.0 | 5.2 | 1/1 | 1.91s | |
| #40 | Qwen3.5-122B-A10B none | Qwen | 10.0 | 5.0 | 1/1 | 2.04s |
| #41 | Qwen3.5-27B none | Qwen | 10.0 | 4.9 | 1/1 | 3.54s |
| #42 | Qwen3.5-35B-A3B none | Qwen | 10.0 | 4.7 | 1/1 | 2.30s |
| #43 | MiniMax M2.5 medium | Minimax | 10.0 | 4.7 | 1/1 | 15.4s |
| #44 | GPT-5.4 none | OpenAI | 10.0 | 4.5 | 1/1 | 2.75s |
| #45 | Trinity Large Preview none | Arcee AI | 10.0 | 4.2 | 1/1 | 6.67s |
| #46 | Kimi K2.5 none | Moonshot AI | 10.0 | 4.1 | 1/1 | 14.0s |
| #47 | GPT-4o-mini none | OpenAI | 10.0 | 4.0 | 1/1 | 2.51s |
| #48 | Qwen3 Coder Next none | Qwen | 10.0 | 4.0 | 1/1 | 2.47s |
| #49 | GLM 4.7 Flash none | Z.ai | 10.0 | 3.9 | 0/1 | 7.05s |
| #50 | Qwen3 Coder Next medium | Qwen | 10.0 | 3.5 | 1/1 | 2.64s |
| #51 | Mercury 2 none | Inception | 10.0 | 3.4 | 1/1 | 1.27s |
| #52 | GLM 4.7 Flash medium | Z.ai | 10.0 | 3.1 | 1/1 | 15.9s |
| #53 | Grok 4.1 Fast none | X AI | 10.0 | 2.9 | 0/1 | 5.51s |
| #54 | MiMo-V2-Flash none | Xiaomi | 10.0 | 2.9 | 1/1 | 2.28s |
| #55 | LFM2-24B-A2B none | Liquid | 10.0 | 2.6 | 0/1 | 0ms |