Expired60% confidenceFactTime unknown
Open-source models such as Qwen-7B and LLaMA-2-7B, after Agent Tuning, can achieve tool-calling accuracy at the level of GPT-3.5 or even close to GPT-4
1
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
6/3/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 9/1/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedOpen-source models such as Llama 3.1 (Meta), Mistral 7B/8x7B, and Qwen2.5-Coder (Alibaba Cloud) have reached practical performance levels approaching GPT-4 on code and configuration generation tasks.80% similarUnverified在文档摘要基准测试上，Qwen2.5-32B-Instruct和Llama-3.1-70B的量化版本已能达到GPT-4早期版本85%-92%的表现水平80% similarUnverifiedQwen2.5-Coder demonstrates performance approaching GPT-4o on code generation benchmarks such as HumanEval, and its 7B version can run on an ordinary laptop after quantization.78% similarUnverifiedQwen2.5-27B是阿里云通义实验室发布的270亿参数开源模型，在MMLU、HumanEval、GSM8K等基准上接近GPT-4早期版本水平77% similarUnverifiedLlama 3提供了8B和70B两个版本，在英文通用任务上的表现已经接近GPT-4级别的闭源模型77% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/40492API
curl https://kongchang.com/api/v1/knowledge/claims/40492MCP
get_claim(id=40492)