Skip to content
Apertis
Get started

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. Llama 3.3 70B InstructMetaRemove
  2. Muse Spark 1.3MetaRemove
  3. Hy4 previewTencentRemove
llama-3.3-70b-instruct vs muse-spark-1.3 vs hy4-preview
AttributeLlama 3.3 70B Instructllama-3.3-70b-instructMuse Spark 1.3muse-spark-1.3Hy4 previewhy4-preview
Pricing
Input$0.10 / 1M$1.25 / 1M$0.834 / 1M
Output$0.25 / 1M$4.25 / 1M$2.50 / 1M
Cache Write (5m)$0.10 / 1M$1.25 / 1M$0.834 / 1M
Cache Write (1h)$0.10 / 1M$1.25 / 1M$0.834 / 1M
Cache Read$0.10 / 1M$1.25 / 1M$0.834 / 1M
Web Search$0 / 1M$0 / 1M$0 / 1M
Context
Max context128K1M1M
Max outputN/AN/AN/A
Capabilities
VisionNoYesYes
Function CallingNoYesYes
JSON ModeNoYesYes
StreamingNoYesYes
Catalogue
ProviderMetaMetaTencent
Categorychatchatchat
Charge typePay As You GoPay As You GoPay As You Go
Released———
Description
Summary—Muse Spark 1.3 is Meta's multimodal reasoning model designed for long-running agentic, multi-agent, and coding workflows. It maintains context and information across extended tasks, enabling reliable execution in complex, multi-step environments. The model is optimized to resolve conflicting information, seek clarification or confirmation when necessary, and execute concisely, making it well suited for autonomous agents, collaborative multi-agent systems, and long-horizon software engineering workflows.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.