ERNIE 5.5 Flash Base Benchmarks: Same-Size SOTA Across Knowledge, Code, Long Context
xiaohu · x · 2026-11-05
Benchmark results for Baidu's ERNIE 5.5 Flash Base (280B-A14) show same-size SOTA across knowledge, reasoning, math, code, multilingual, and long-context suites, beating DeepSeek-V4-Flash-Base, Hy3-preview-Base, and MiMo-V2.5-Base under the same evaluation pipeline.
Standouts include MMLU 93.3, BBH (CoT) 89.2, HumanEval+ 82.3 Pass@1, and RepoQA 91.7 Pass@1.
More from Models
- Unsloth Ships Dynamic V3.1 Quants Claiming 10-15% Accuracy Gains, Reddit Skeptical — fiocalisti · 2026-11-12
- KRAFTON Open-Sources Raon-1 nano: 1.5B Effective Params, 262K Context, On-Device Ready — pmttyji · 2026-11-12
- Robot build test: GPT-6 Astra stabilizes two-board structure, Claude Fable 5.2 falls short — qinzytech · 2026-11-12
- Zvi dissects Anthropic's latest system card, disputing the autonomy threshold call — Don't Worry About the Vase (Zvi) · 2026-11-12
- antirez backs low-bit DeepSeek 4.1 Flash over Qwen 27B dense on Apple M5 Macs — antirez · 2026-11-12
- Gemini re-renders images as text when vision is removed, dropping only 6% WSR — zacharynado · 2026-11-12