spacehub

surshar / claude-code

Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.

Built by Anthropic · github.com/anthropics

claude-code / plugins / claude-opus-4-5-migration / FAST_MODEL_CONFIG.md
70 lines · 2.0 KBPreviewCodeRaw
1# Fast Opus 4.5 Model Configuration
2
3## Issue
4The standard Opus 4.5 configuration is limited to 200,000 tokens, causing failures with larger contexts.
5
6## Solution: Fast Opus Deployment
7
8### Environment Variable Configuration
9
10To use the fast Opus deployment with extended context length, configure Claude Code with:
11
12```bash
13# Set the Opus alias to use the fast deployment
14export ANTHROPIC_DEFAULT_OPUS_MODEL="claude-opus-4-5-20251101-fast"
15```
16
17### Model String Updates for Fast Deployment
18
19| Platform | Fast Opus 4.5 Model String |
20|----------|---------------------------|
21| Anthropic API (1P) | `claude-opus-4-5-20251101-fast` |
22| AWS Bedrock | `anthropic.claude-opus-4-5-20251101-fast-v1:0` |
23| Google Vertex AI | `claude-opus-4-5-fast@20251101` |
24| Azure AI Foundry | `claude-opus-4-5-20251101-fast` |
25
26### Context Length Configuration
27
28The fast deployment supports extended context. Configure the context limit:
29
30```bash
31# Extended context configuration (up to 1M tokens)
32export ANTHROPIC_OPUS_CONTEXT_LIMIT="1000000"
33```
34
35### Integration with Migration Tool
36
37When migrating to Opus 4.5 with extended context requirements:
38
391. Use the fast model strings above instead of standard ones
402. Set the environment variable for Opus alias
413. Configure extended context limit
424. Test with large contexts (>200k tokens)
43
44### VSCode Configuration
45
46For VSCode extension users:
47
48```json
49{
50 "claude-code.environmentVariables": {
51 "ANTHROPIC_DEFAULT_OPUS_MODEL": "claude-opus-4-5-20251101-fast",
52 "ANTHROPIC_OPUS_CONTEXT_LIMIT": "1000000"
53 }
54}
55```
56
57### Verification
58
59Test the configuration with a prompt that exceeds 200k tokens to ensure the error:
60```
61prompt is too long: 282789 tokens > 200000 maximum
62```
63is resolved.
64
65## When to Use Fast vs Standard
66
67- **Use Fast**: For large codebases, extensive context analysis, long conversations
68- **Use Standard**: For typical development tasks with smaller context requirements
69
70The fast deployment trades some response speed for much higher context capacity.