4 models
Ranked by DeepSWE
Reasoning Tools JSON 71.0

xAI's frontier model for long-running agents, coding, knowledge work, and visual projects

xai/grok-4.7 2026-09-21 500K context $2/M input $6/M output
27 providers
Reasoning Tools JSON 67.2

Cost-efficient GPT-5.6 model for fast, high-volume workloads

openai/gpt-5.6-luna 2026-07-09 1.05M context $0.2/M input $1.2/M output
45 providers
Reasoning Tools JSON 66.6

OpenAI's most efficient model for focused, high-volume tasks

openai/gpt-6-luna 2026-09-22 1.05M context $0.1/M input $0.5/M output
35 providers
Reasoning Tools Open weights 63.4

Native multimodal GLM model for efficient coding and long-horizon agent tasks

zhipuai/glm-5.3-flash 2026-08-26 1M context $0.12/M input $0.418/M output
65 providers