Unverified50% confidenceOpinionExact time
The 27.18 MiB cache size makes it possible for Mamba2-8B to run on high-end mobile devices or even embedded systems.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/16/2026
First Seen
Valid until: 1/14/2027
Sources
Mamba2-8B Compression Research: 27MB Cache and 3.14GB Model
redditr/deeplearning10/16/2026
Related Entities
Related Claims
UnverifiedDeduplicating the Q/K/V weights saved 488 MiB of VRAM, and keeping the text embedding table on the CPU saved 594 MiB, enabling the two models Gemma 4 E4B and Qwen3-TTS 1.7B to be loaded simultaneously, with the graphics card showing 7,448 MiB used out of 8,188 MiB.71% similarUnverifiedA mid-range machine equipped with a 5060Ti graphics card, 16GB of VRAM, and 32GB DDR4 RAM can fully run the ComfyUI workflow collection combining Qwen3 and MiniMax H3.67% similarUnverifiedA user runs the 397B model in FP8 on 8× GB10 units at a speed 20% faster than INT4, and published the first solution for MiMo 2.5 Pro and Kimi 2.6 on the NVIDIA forums.66% similarUnverifiedThe 4-bit quantized weights of Qwen3 27B are only approximately 16.1 GB, allowing it to run on a MacBook Pro/Mac Studio with 32 GB of memory or a consumer-grade GPU with approximately 17 GB of VRAM.65% similarUnverifiedThis build uses the Mingxuan AI Micro Station workstation barebone, equipped with an Intel Ultra 9 275HX (24-core), 96GB of RAM, two 4TB M.2 drives, and two ARC Pro B70 cards.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1048603API
curl https://kongchang.com/api/v1/knowledge/claims/1048603MCP
get_claim(id=1048603)