OpenCodePapers

instruction-following-on-multi-if

Natural Language UnderstandingInstruction Following
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
PaperCodeAccuracyModelNameReleaseDate
ERNIE 5.0 Technical Report85.56ERNIE 5.02026-02-04
LFM2.5-24B-A2B: Built for Fast On-Device Agents84.38LFM2.5-24B-A2B2026-10-13
LFM2.5-2.6B: Deploy Agents Everywhere✓ Link80.07LFM2.5-2.6B2026-08-04
LFM2.5-8B-A1B: an Even Better on-Device Mixture-of-Experts✓ Link79.93LFM2.5-8B-A1B2026-05-28
Qwen3-VL Technical Report✓ Link79.1Qwen3-VL (235B-A22B, Thinking)2025-11-26
ERNIE 4.5 Technical Report✓ Link76.6ERNIE-4.5-300B-A47B2025-06-30
Qwen3 Technical Report✓ Link71.9Qwen3-235B-A22B (Thinking)2025-05-14
Data Science and Technology Towards AGI Part I: Tiered Data Management✓ Link71.8MiniCPM5-2B2026-02-09
Qwen3 Technical Report✓ Link70.2Qwen3-235B-A22B (Non-thinking)2025-05-14
LFM2.5-1.2B-Thinking: On-Device Reasoning Under 1GB69.33LFM2.5-1.2B-Thinking2026-01-20
SEIF: Self-Evolving Reinforcement Learning for Instruction Following✓ Link62.4SEIF (Llama-3.1-8B-Instruct)2026-05-08
Introducing LFM2.5: The Next Generation of On-Device AI60.98LFM2.5-1.2B-Instruct2026-01-05
LFM2 Technical Report✓ Link58.19LFM2-8B-A1B2025-11-28
LFM2.5-350M: No Size Left Behind44.92LFM2.5-350M2026-03-31
MiniCPM5-1B✓ Link43.54MiniCPM5-1B (Thinking)2026-05-25
LFM2.5-230M: Built to Run Anywhere37.70LFM2.5-230M2026-06-25
LFM2.5-VL-450M: Structured Visual Intelligence, Edge to Cloud34.63LFM2.5-VL-450M2026-04-08