Developer
Baidu
8 indexed models; 1 currently meet the evidence threshold for the overall ranking. Together they have results from 13 evaluations.
Models by Baidu
| Model | Evals | Components | Rank | Release date |
|---|---|---|---|---|
| Ernie 4.5 300B A47B | 6 | 6/7 | 132 | 2025-06-28 |
| Ernie 3.5 128K | 1 | 1/7 | — | — |
| Ernie 4.5 21B A3B | 1 | 4/7 | — | 2025-09-08 |
| ERNIE 4.5 VL 424B A47B | 1 | 1/7 | — | 2025-06-28 |
| Ernie 5.0 Preview | 2 | 4/7 | — | 2025-10-22 |
| Ernie 5.0 Thinking Preview | 1 | 1/7 | — | — |
| Ernie 5.1 | 2 | 1/7 | — | — |
| Ernie Bot | 2 | 3/7 | — | 2023-03-16 |
Evaluations covering Baidu models (13)
AA-Omniscience · AgentDrive Safety Compliance · Arena Factuality — Search Arena (factuality-only weighting) · Arena Factuality — Text Arena (factuality-only weighting) · BullshitBench v2 · Confabulations · FlagEval Safety and Values · FLAMES · LiveSecBench · M3-SafetyBench · Opposite-Narrator Sycophancy · S-Eval · UAVBench safety-critical decision recognition