Four models, led by Opus
OpusVIP is tuned for Opus. Opus 5.5 is the flagship workhorse, Fable 5.1 takes on the hardest reasoning and long-horizon agent work, and Sonnet 5.5 and Haiku 5.5 are strong partners for subagents and light tasks in Claude Code. All four are billed at 1x Anthropic list price and converted to credits at 100 credits per $1 of usage ($0.15 to you). Switch any time with /model.
$0.15 buys 100 credits ($1 of usage at list price), so you pay 15%.
Last updated:
Claude Opus 5.5
Recommendedclaude-opus-5-5 Flagship workhorse: complex coding, architecture, long tasks and hard bugs. 1M context.
- Input (credits / 1M)
- 400
- You pay $0.60 · List $4
- Output (credits / 1M)
- 2,000
- You pay $3.00 · List $20
Claude Fable 5.1
claude-fable-5-1 Anthropic's most capable model: the hardest reasoning and long-horizon agent work. 1M context.
- Input (credits / 1M)
- 1,000
- You pay $1.50 · List $10
- Output (credits / 1M)
- 5,000
- You pay $7.50 · List $50
Claude Sonnet 5.5
claude-sonnet-5-5 Balanced partner: subagents, bulk edits and everyday development. 1M context.
- Input (credits / 1M)
- 200
- You pay $0.30 · List $2
- Output (credits / 1M)
- 1,000
- You pay $1.50 · List $10
Claude Haiku 5.5
claude-haiku-5-5 Fast and light: small tasks, session titles and quick questions. 1M context.
- Input (credits / 1M)
- 10
- You pay $0.015 · List $0.10
- Output (credits / 1M)
- 50
- You pay $0.075 · List $0.50
Prompts over 100K tokens use Anthropic's long-context rate: input 50 · output 250 (credits / 1M)
Price comparison
Fable 5.1 is Anthropic's most capable model, Sonnet 5.5 costs half as much as Opus 5.5, and Haiku 5.5 costs 1/40 as much. Cache prices follow Anthropic's ratios for every model: a 5-minute write is 1.25x the input price, a 1-hour write is 2x, and a cache read costs far less than regular input. For each model's capabilities, see Anthropic's documentation.
| Model | Context | Input | Output | Cache read | Cache write 5 min | Cache write 1 h |
|---|---|---|---|---|---|---|
| Claude Opus 5.5 | 1M | 400 | 2,000 | 20 | 500 | 800 |
| Claude Fable 5.1 | 1M | 1,000 | 5,000 | 25 | 1,250 | 2,000 |
| Claude Sonnet 5.5 | 1M | 200 | 1,000 | 10 | 250 | 400 |
| Claude Haiku 5.5 | 1M | 10 | 50 | 1 | 12.5 | 20 |
Unit: credits per 1M tokens (1 credit = $0.01 at list price; 100 credits = $0.15), billed at 1x list price. Haiku 5.5 prompts over 100K tokens use Anthropic's long-context rate: input 50, output 250, cache read 5, and cache write 62.5 (5 min) or 100 (1 h) credits.
Switching in Claude Code
Claude Code's default aliases map straight to these four models: opus is Opus 5.5, fable is Fable 5.1, sonnet is Sonnet 5.5 and haiku is Haiku 5.5. Switch with the --model flag or the in-session /model command. No extra configuration is needed; see Choose and switch models.
How to choose: give the main session to Opus 5.5, send the hardest reasoning and long-horizon work to Fable 5.1, use Sonnet 5.5 for subagents, bulk edits and everyday development, and use Haiku 5.5 for light tasks and quick questions. More in Opus 5.5, Fable 5.1, Sonnet 5.5 or Haiku 5.5.
# Pick at startup
claude --model opus
# Switch inside a session
/model sonnet
/model fable
/model haiku