Skip to content

Compare models

Put up to 4 models beside each other — token prices, context windows, capabilities and provider, from the same catalogue the model pages read.

  1. Gemini 2.5 Flash DeepSearchGoogleRemove
  2. Hy4 previewTencentRemove
  3. GLM 5.3 FlashZ.AIRemove
gemini-2.5-flash-deepsearch vs hy4-preview vs glm-5.3-flash
AttributeGemini 2.5 Flash DeepSearchgemini-2.5-flash-deepsearchHy4 previewhy4-previewGLM 5.3 Flashglm-5.3-flash
Pricing
Input$4.20 / 1M$0.834 / 1M$0.075 / 1M
Output$33.60 / 1M$2.50 / 1M$0.25 / 1M
Cache Write (5m)$4.20 / 1M$0.834 / 1M$0.075 / 1M
Cache Write (1h)$4.20 / 1M$0.834 / 1M$0.075 / 1M
Cache Read$4.20 / 1M$0.834 / 1M$0.075 / 1M
Web Search$0 / 1M$0 / 1M$0 / 1M
Context
Max context1.0M1M1M
Max outputN/AN/AN/A
Capabilities
VisionYesYesYes
Function CallingYesYesYes
JSON ModeYesYesYes
StreamingYesYesYes
Catalogue
ProviderGoogleTencentZ.AI
Categorychatchatchat
Charge typePay As You GoPay As You GoPay As You Go
Released
Description
SummaryGemini 2.5 Flash is Google's main high-performance model for complex reasoning, coding, math, and scientific tasks. It has built-in “thinking” features that help it produce more accurate, context-aware answers.Tencent Hy4 Preview is a Mixture-of-Experts (MoE) model from Tencent, featuring 770B total parameters with 49B activated per token. It is designed for coding agents, complex tool-driven workflows, and professional productivity tasks that require strong planning and reliable execution. Optimized for context continuity and sustained multi-step work, Hy4 Preview is well suited for long-horizon coding, agentic automation, tool orchestration, and complex real-world workflows.GLM-5.3-Flash is Z.AI's efficient native multimodal model, designed for coding and long-horizon agentic workflows. It combines strong multimodal capabilities with an architecture optimized for responsive, cost-efficient task execution. Built on a hybrid sparse and linear attention architecture, GLM-5.3-Flash maintains accurate long-context behavior while reducing computational overhead, making it well suited for coding agents, extended multi-step tasks, and scalable production workloads.