Claude Fable 5 vs Claude Fable 5.1
The two are level across all three aggregates. Every numeric field is compared below, with a worked monthly-cost example and a pick rule for each use case.
Verdict at a glance
Choose Claude Fable 5 if you…
- Strongest all-round reasoning
- Reliable long-context recall
- Top-tier coding
- Best for: Deep reasoning, Long documents, Coding
Choose Claude Fable 5.1 if you…
- Top agentic coding (Terminal-Bench 55.8)
- Highest AA Intelligence Index
- Reliable 1M long-running agents
- Cache reads 75% cheaper
- Best for: Agentic coding, Long-horizon tasks, Scientific research
Head-to-head aggregate scores
Scores are 0–100, aggregated from public benchmark information and independently weighted across three leaderboards. Rank is out of 26 tracked models.
Aggregated from public sources and independently weighted; methodology on the Terms page. Scores within 3 points are treated as statistically tied.
Specs & pricing — every field side by side
List API prices in USD per 1M tokens. The highlighted cell is the stronger value/capability on that row.
| Dimension | Claude Fable 5 | Claude Fable 5.1 | Verdict |
|---|---|---|---|
| Vendor | Anthropic (US) | Anthropic (US) | Same vendor |
| Released | 2026.06 | 2026.09 | Claude Fable 5.1 is newer |
| Overall (rank) | 95 · #2 | 95 · #2 | Tie |
| Coding | 99 · #1 | 99 · #1 | Tie |
| Multimodal | 97 · #1 | 97 · #1 | Tie |
| Context window | 1M | 1M | Tie |
| Max output | 128K | 128K | Tie |
| Effective-context | 98 | 99 | Claude Fable 5.1 more reliable |
| Input $/1M | $10 | $10 | Tie |
| Output $/1M | $50 | $50 | Tie |
| Cache discount | 90% off | 90% off | Tie |
| Speed | ~30 tok/s | ~32 tok/s | Claude Fable 5.1 faster |
| TTFT | 1.2s | 1.1s | Claude Fable 5.1 snappier |
| Function calling | 92 | 93 | Claude Fable 5.1 ahead |
| Refusal rate | ~8% | ~9% | Claude Fable 5 less restrictive |
| English | 98 | 99 | Claude Fable 5.1 |
| Chinese | 82 | 83 | Claude Fable 5.1 |
| Modalities | text, image | text, image | Same |
| Open weights | No | No | Both closed |
| Fine-tuning | No | No | |
| Free tier | No free tier; $5 credit to start | No free tier; cache reads cut to a quarter of cost | |
| SOC2 / no-train | yes / yes | yes / yes | |
| Private deployment | No | No | Neither |
Fields drawn from vendor public documentation and the Modelspectra 26-model dataset; speed varies with network, concurrency and prompt length. Verify current pricing before purchase.
Dimension-by-dimension analysis
Reasoning & overall intelligence
The two are level on the overall aggregate (95/100 each), a gap inside the 3-point band where rankings are statistically indistinguishable and task-specific results can swap.
Agentic coding
Both score 99/100 on the coding aggregate, so expect parity on most day-to-day engineering tasks.
Multimodal
They tie on the multimodal aggregate at 97. Neither emits native video, so the comparison is about parsing images and documents, not generation.
Speed & latency
Claude Fable 5.1 is faster in interactive use: ~32 tok/s with 1.1s TTFT versus ~30 tok/s with 1.2s TTFT (about 1.1× the throughput). For conversational UIs where perceived responsiveness drives retention, that edge is a real product factor even when raw reasoning is lower.
Context: window vs usable recall
Nominal windows are 1M for Claude Fable 5 and 1M for Claude Fable 5.1. Crucially, the larger nominal window does not win on usable recall: Claude Fable 5 advertises 1M but Claude Fable 5.1 scores higher on effective-context (99 vs 98), i.e. it actually retains more of what it was given.
Price & total cost
Claude Fable 5 is the cheaper API at $10/$50 versus Claude Fable 5.1 at $10/$50 per 1M input/output tokens — list input is about 1.0× lower. Cache discounts (Claude Fable 5 90% vs Claude Fable 5.1 90%) shift the effective bill, worked out below.
Chinese vs English
English: Claude Fable 5 98 vs Claude Fable 5.1 99. Chinese: 82 vs 83. Both are US-based models; for Chinese-first workloads also compare domestic models on the leaderboard.
Tool use & ecosystem
Function-calling score: Claude Fable 5 92 vs Claude Fable 5.1 93, so Claude Fable 5.1 has the edge on structured tool use. Neither offers standard fine-tuning. Factor in existing SDK/plugin familiarity — switching cost often outweighs a few-point tool-use gap.
Cost worked example — same workload, real token math
Assume a production workload of 100M input + 30M output tokens per month, with 90% of input tokens served from cache. Figures use public list prices.
| Scenario · per month | Claude Fable 5 | Claude Fable 5.1 | Gap |
|---|---|---|---|
| List priceno cache applied | $2,500100M in × $10 + 30M out × $50 | $2,500100M in × $10 + 30M out × $50 | 1.00×even |
| With caching90% of inputs cache-hit | $1,69090M cached in × $1 + 10M in × $10 + 30M out × $50 | $1,69090M cached in × $1 + 10M in × $10 + 30M out × $50 | 1.00×even |
Illustrative model; your input/output mix and cache-hit ratio change the result. Prices are list rates before any enterprise agreement.