Teamday

Chat & Reasoning Model · Anthropic

Claude Opus 4.8

Claude Opus 4.8 is a previous-generation Anthropic Opus model for the hardest coding, long-running agents, and complex multi-step reasoning. It was succeeded by Claude Opus 5 and then Claude Opus 5.5 ($4 / $20 per 1M tokens); Teamday agents configured with Opus 4.8 now run on Claude Opus 5.5.

Use Claude Opus 4.8 in Teamday →All models

At a glance (verified 2026-09-05)

Vendor
Anthropic
Type
Chat & Reasoning Model
Pricing
Base: $5 in / $0.50 cache hits / $25 out per 1M tokens. Cache writes: $6.25 5m / $10 1h.

Best for

Hardest coding problemsLong-running agentsComplex multi-step reasoningSecurity review

How to use Claude Opus 4.8 in Teamday

Claude Opus 4.8 is no longer in Teamday's chat model picker — pick Claude Opus 5.5, its current successor on the Claude Code harness. Existing agents set to Opus 4.8 are routed to Opus 5.5 automatically.

Harness: claude-code →Max →James →

Frequently asked questions

How much does Claude Opus 4.8 cost per million tokens?

Base pricing is $5 per 1M input tokens, $0.50 for cache hits, and $25 per 1M output tokens; cache writes cost $6.25 (5-minute) or $10 (1-hour).

When should I pick Claude Opus 4.8 over Fable 5.1?

You no longer need to — Claude Opus 5.5 now holds the Opus slot at $4 vs Fable 5.1's $10 per 1M input tokens. The trade-off is unchanged: Opus for most coding and agent work, Fable 5.1 for the most demanding long-horizon work.

Is Claude Opus 4.8 available in Teamday?

It has been replaced by Claude Opus 5.5 in Teamday's chat model picker. Agents and chats previously set to Opus 4.8 run on Claude Opus 5.5 automatically, at a lower price.

Related

Claude Opus 4.8 is one of the models Teamday's AI employees already run — no API keys, no infrastructure.

Hire your first AI employee →