Compare models
Pick two to four models. No markup on model usage: you pay the provider's list price.
| Attribute |   Claude Opus 5.5 |  Qwen3.8 Max |
|---|---|---|
| Provider | Anthropic | Alibaba Cloud (Qwen) |
| Model id | anthropic/claude-opus-5-5 | qwen/qwen3.8-max |
| Status | Available via Safeguard | Available via Safeguard |
| Input / 1M tokens | $4 | $2 |
| Output / 1M tokens | $20 | $6 |
| Price notes | Base price, full 1M context at standard rate. Cache hits $0.20/1M. Batch 50% off. US-only inference 1.1x. | International (Singapore) deployment, 0 to 1M input tokens, both modes. |
| Context | 1M | 1M |
| Parameters | Not published | Not published |
| Max output | 128,000 tokens | Not published |
| Modality | Text and image | Text and image |
| Weights | Closed | Closed |
| Zero data retention | Available | Not stated |
| Prompts go to | Anthropic (global routing by default) | Alibaba Cloud Model Studio (prices shown are for the Singapore / International deployment) |
| Retention | Not used for model training without express permission, per Anthropic's API retention doc, conversation content is not retained by default except for Covered Models that require 30-day retention, flagged content and legal holds can be kept longer | Alibaba Cloud states it never uses your data for model training, Model Studio stores data generated from model calls as required by law (no period stated) |
| Region | Global by default, US-only inference via inference_geo at 1.1x pricing (Claude 4.6 and later) | Singapore (International) for these prices, other regions such as China (Beijing), Hong Kong, Frankfurt, Virginia and Tokyo have separate price lists |
| Price checked | 2026-10-08 | 2026-10-08 |
Self-healing security runs on Safeguard.
Your first fix PR is minutes away.
No sales call required, even your agent can complete the purchase over MCP.