SWE-bench Verified
Read the Portuguese translation →
Mentioned in
Section titled “Mentioned in”- Flint 40B vs Qwen3.8 27B
- This Tiny Coding Model Shouldn’t Be This Good (MiniCPM5)
- LFM2.5 24B-A2B: Did Liquid AI Manage to Beat Qwen? Tested Locally
- Qwen 27B on 6GB VRAM…
- Run Qwen 3.8 27B in 8GBs Saluki 27B
- Saluki 27B: a 54 GB model squeezed into 7.89 GB (88 vs 84 on tool calling)
Related
Section titled “Related”llama.cpp, Qwen3.8 27B, Terminal-Bench 2.1, MLX, GGUF, Berkeley function calling leaderboard, Saluki 27B, Hugging Face, Underdog, Recursive, Mixture of Experts, Flint