AI BENCHY Category Failures
Data parsing and extraction: API error
Data parsing and extraction
API error
See which AI models are most likely to hit API error on Data parsing and extraction, so you can spot weak points faster. Sort by: Response Time (avg) ↑.
Failure Reasons
| Rank | Model | Company | API error Count | Category Score | Tests Correct | Response Time (avg) |
|---|---|---|---|---|---|---|
| #64 | MiMo-V2-Flash medium | Xiaomi | 1 | 6.5 | 1/2 | 0ms |
| #83 | Step 3.5 Flash none | Stepfun | 1 | 3.0 | 0/1 | 0ms |
| #132 | Mistral Small 4 medium | Mistral | 1 | 7.3 | 1/2 | 1.23s |
| #156 | Hy3 preview none | Tencent | 1 | 6.5 | 1/2 | 2.85s |
| #33 | Hy3 preview medium | Tencent | 1 | 6.5 | 1/2 | 5.25s |
| #89 | Hy3 preview low | Tencent | 1 | 6.5 | 1/2 | 5.85s |
| #126 | gpt-oss-120b none | OpenAI | 1 | 6.5 | 1/2 | 7.12s |
| #20 | Gemini 3.5 Flash none | 1 | 6.5 | 1/2 | 8.10s | |
| #100 | Grok Build 0.1 none | X AI | 1 | 3.8 | 0/2 | 9.33s |
| #82 | Hy3 preview high | Tencent | 1 | 6.5 | 1/2 | 12.1s |
| #152 | MiMo-V2-Flash none | Xiaomi | 1 | 2.9 | 0/2 | 19.7s |
| #103 | DeepSeek V4 Pro high | DeepSeek | 1 | 7.3 | 1/2 | 23.6s |
| #113 | DeepSeek V4 Pro none | DeepSeek | 1 | 6.9 | 1/2 | 30.5s |
| #96 | Ring-2.6-1T none | Inclusionai | 1 | 3.0 | 0/2 | 45.9s |
| #49 | Qwen3.5-Flash medium | Qwen | 1 | 7.3 | 1/2 | 57.0s |