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.
| Attribute | GLM 5.3 Flashglm-5.3-flash | Seedream 3.0doubao-seedream-3-0-t2i-250415 | Hy4 previewhy4-preview |
|---|---|---|---|
| Pricing | |||
| Input | $0.075 / 1M | — | $0.834 / 1M |
| Output | $0.25 / 1M | — | $2.50 / 1M |
| Cache Write (5m) | $0.075 / 1M | Not applicable | $0.834 / 1M |
| Cache Write (1h) | $0.075 / 1M | Not applicable | $0.834 / 1M |
| Cache Read | $0.075 / 1M | Not applicable | $0.834 / 1M |
| Web Search | $0 / 1M | — | $0 / 1M |
| Request | — | $0.075 / request | — |
| Billing | — | Pay Per Request | — |
| Context | |||
| Max context | 1M | 128K | 1M |
| Max output | N/A | N/A | N/A |
| Capabilities | |||
| Vision | Yes | Yes | Yes |
| Function Calling | Yes | No | Yes |
| JSON Mode | Yes | No | Yes |
| Streaming | Yes | No | Yes |
| Catalogue | |||
| Provider | Z.AI | ByteDance | Tencent |
| Category | chat | image | chat |
| Charge type | Pay As You Go | Pay Per Request | Pay As You Go |
| Released | — | — | — |
| Description | |||
| Summary | 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. | Doubao-Seedream-3.0-T2I-250415 is ByteDance's advanced text-to-image (T2I) generation model, designed to deliver high-quality visual outputs with strong prompt alignment and fast inference performance. It supports detailed scene composition, rich stylistic control, and high-fidelity rendering across a wide range of creative use cases. Optimized for production deployment, Seedream 3.0 balances visual quality, generation speed, and cost efficiency, making it well suited for creative content generation, design workflows, marketing assets, and interactive image-based applications. | 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. |