Claude Fable 5
API requires an overseas network; private deployment is negotiable through the enterprise channel.
01
Snapshot
Overall
95
rank #1 of 22
Coding
99
rank #1 of 22
Multimodal
97
rank #1 of 22
Input / Output
$10 / $50
USD per 1M tokens
Context
1M
usable recall 98
Speed
~30 tok/s
TTFT 1.2s
Aggregated from public sources and independently weighted; methodology on the Terms page. Scores within 3 points are treated as statistically tied.
02
Specs & pricing
Every tracked field. List API prices in USD per 1M tokens; verify current vendor pricing before purchase.
| Vendor | Anthropic (US) |
| Released / current | 2026.06 |
| Licensing | Closed / managed API |
| Context window | 1M |
| Max output | 128K |
| Effective-context score | 98 |
| Input / output price | $10 / $50 per 1M tokens |
| Cache discount | Prompt cache 90% off |
| Free tier | No free tier; $5 credit to start |
| Speed / TTFT | ~30 tok/s / 1.2s |
| Function calling | 92 |
| Refusal rate | ~8% |
| English / Chinese | 98 / 82 |
| Modalities | text, image |
| Fine-tuning | No |
| Private deployment | No |
| SOC2 / no-train | yes / yes |
03
Strengths & watch-outs
Strengths
- Strongest all-round reasoning
- Reliable long-context recall
- Top-tier coding
Watch-outs
- Highest price
- Relatively slow
- No native video
04
Where it ranks among 22 models
Overall 95/100 (#1), coding 99 (#1), multimodal 97 (#1). See the full boards and side-by-side compare on the leaderboard.
05
Pricing in practice
Illustrative monthly bill for 100M input + 30M output tokens: $2,500 at list; about $1,690 with 90% of inputs cache-hit. Your mix and cache-hit ratio change this.
Illustrative model using public list rates before any enterprise agreement; cache discount: Prompt cache 90% off.
06
Frequently asked questions
How much does Claude Fable 5 cost?
List API pricing is $10 input and $50 output per 1M tokens. Prompt cache 90% off. No free tier; $5 credit to start
What is Claude Fable 5's context window, and is it usable end to end?
It advertises a 1M window with 128K max output; its effective-context score is 98/100, which is the better predictor of whether long-document details are actually retained.
Is Claude Fable 5 good for coding?
Its coding aggregate is 99/100, rank #1 of 22. That puts it in the top tier for agentic, multi-file engineering.
Can I self-host Claude Fable 5?
No. It is a closed managed API; there is no weights download.
How strong is it in Chinese and on multimodal inputs?
Chinese score is 82 versus English 98; multimodal aggregate is 97/100 (rank #1), accepting text, image.