@jinyuhou0: On popular benchmarks, our 30B model matches systems 20-30x its size (gpt-5.4-xhigh, DeepSeek-V3.2, Kimi-K2.5), while u…
Summary
A new 30B model matches systems 20-30x its size on popular benchmarks while using up to 95% fewer reasoning tokens than comparable agentic LLMs, achieved through a learned configurator that decides when and how to reason. Model and code are openly available.
View Cached Full Text
Cached at: 05/24/26, 10:27 AM
On popular benchmarks, our 30B model matches systems 20-30x its size (gpt-5.4-xhigh, DeepSeek-V3.2, Kimi-K2.5), while using up to 95% fewer reasoning tokens than comparable 30/32B agentic LLMs.
The trick: don’t just reason less, reason about the right things. A learned configurator decides when to simulate, how far ahead, and when to skip planning entirely.
Efficient reasoning is an allocation problem, not a compression problem.
Model and code are openly available.
Mingkai Deng (@mdeng34): Frontier LLMs are converging on efficient, adaptive reasoning. Opus 4.7 lets the model decide how deeply to reason. GPT-5.5 achieves strong results with fewer reasoning tokens.
We study a related but more structural question: what 𝗸𝗶𝗻𝗱 𝗼𝗳 𝗿𝗲𝗮𝘀𝗼𝗻𝗶𝗻𝗴 should we
Similar Articles
@omarsar0: Small models are getting really good at reasoning. It's exciting because SLMs can unlock so much at the harness layer. …
TheWebAI released TwIL-LM3-Pro, a 3.6B-parameter open-source SLM that runs locally and achieves 95.4 on BIG-Bench Hard, outperforming Qwen3-8B's 63.7, using a post-training recipe of formal-logic fine-tuning, weight merging, and RL with a programmatic verifier.
@rohanpaul_ai: China's best small reasoning model (VibeThinker-3B) just got beaten by a 3.6B model from Austin. > webAI's TwIL-LM3-Pro…
webAI released TwIL-LM3-Pro, a 3.66B open-source reasoning model that surpasses VibeThinker-3B, Qwen3.5-4B, and LFM2.5-8B-A1B on formal logic benchmarks and runs locally via llama.cpp in a ~2 GiB Q4 build.
@akshay_pachaar: Chinese researchers did it again! OpenBMB just open-sourced MiniCPM5-2B, a dense 2B-parameter model built for reasoning…
OpenBMB has open-sourced MiniCPM5-2B, a 2B-parameter AI model optimized for reasoning, coding, and tool use on resource-constrained hardware, achieving state-of-the-art performance in its size class and demonstrating effective local deployment capabilities.
We built our own inference engine so a 32B model could keep 262K context inside a 16 GB GPU budget. Here's an uncut run.
Blockway 团队发布了 Volundr 32B 模型及自研 Agens Engine 推理引擎（MIT 开源），通过让 72 层中的 54 层使用固定大小状态替代增长的 KV cache，使 262K 长上下文能在 16 GB GPU 预算内运行，并演示了 110K token 冷读约 69 秒、后续追问 0.5 秒回答的实测。
LFM3-1.7b seems impressive
The author tested Liquid AI's LFM3-1.7B model (Q8_0 GGUF via llama.cpp/ROCm) for compression and summarization, reporting strong prefill/decode throughput on Strix Halo and tensor-split 7900 XTX+XT setups, plus unexpectedly coherent performance at 32k-64k context lengths for a 1.7B model.