Expired50% confidenceBenchmarkExact time
In a tower defense game construction test, the 35B model generated approximately 750 lines of code with a basically playable game, while the 9B model generated nearly 900 lines of code but the game was almost unplayable.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/7/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 10/5/2026(expired)
Sources
Ornith 9B Tested: Can a 16GB Mac Mini Handle Local AI Coding?
bilibili四月的尾巴_7/4/2026
Related Claims
UnverifiedA tester ran a 20.7B dense model on a dual-4090 (48 GB VRAM) workstation and achieved 27.6 tokens per second, only slightly faster than Dwarf Star, but with over 10 times fewer parameters.64% similarUnverified受当时硬件限制，《魔兽争霸III》游戏角色模型通常仅由数百至数千个多边形构成，贴图分辨率普遍在256×256像素以下60% similarUnverifiedUsers with an NVIDIA RTX 4090 (24 GB VRAM) can run 70B-parameter models locally for personal experimentation and development via 4-bit quantized versions in GGUF format.59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/176267API
curl https://kongchang.com/api/v1/knowledge/claims/176267MCP
get_claim(id=176267)