Expired50% confidenceBenchmarkExact time
7–8B is the minimum threshold for lx tools; llama3.1:8b and qwen2.5:7b perform well, and 14B-class models can basically achieve cloud API-level results.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/20/2026
First Seen
Valid until: 10/18/2026(expired)
Sources
Related Claims
Unverifiedllama3.1:8b and qwen2.5:7b perform well on lx, and 7–8B is the minimum threshold for this toolset to function properly.85% similarUnverifiedLFM2.5-8B-A1B总参数量为8B，推理时仅激活1.5B参数67% similarUnverifiedIn MiniMax's MMClaw evaluation, M2.7 approaches the level of Sonnet 4.6 in open-ended tool-calling scenarios65% similarVerifiedQwen 2.5 is available in multiple parameter versions including 1.5B, 7B, and 72B.65% similarUnverified在HumanEval、MBPP、GSM8K等基准测试上，Gemma 7B的表现显著优于同参数量级的LLaMA 2 7B和Mistral 7B65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/573254API
curl https://kongchang.com/api/v1/knowledge/claims/573254MCP
get_claim(id=573254)