Claude Fable 5.1 benchmarks

Claude Fable 5.1 is a proprietary model from Anthropic, released 1 Sep 2026. It ranks 3rd of 51 on our overall leaderboard and 1st of 42 for coding. At A$28.08 per million tokens (blended), it costs 7.8× the median model we track. It writes about 65 tokens a second, and answers start after 2.0 min on average, thinking included. It can be called from AWS Bedrock in Sydney and Melbourne and Google Vertex AI in Sydney, but only through global routing, so requests may be processed outside Australia.

#3 OF 51 OVERALLA$28.08 / 1M TOKENS1M CONTEXTUPDATED 23 SEP 2026
CONTEXT WINDOW
1,000,000 tokens
MAX OUTPUT
128,000 tokens
INPUT
text, image
WEIGHTS
Proprietary
RELEASED
1 Sep 2026
OUTPUT SPEED
65 tok/s
ANSWER STARTS AFTER
2.0 min (thinking included)
ANTHROPIC DOCS
§ 02 — RESULTS

Benchmark results.

Independent test scores are for the adaptive reasoning, max effort setting. “Among tracked” ranks Claude Fable 5.1 against the 51 models on this site; LMArena ranks run across its full leaderboard.

ALL RESULTS8 OF 8 CORE SIGNALS
Claude Fable 5.1 benchmark results
BENCHMARKRESULTAMONG TRACKEDSOURCE DETAIL
INDEPENDENT TESTS · ARTIFICIAL ANALYSIS
AA Intelligence Index53.4#2 of 51Adaptive reasoning, max effort
AA Coding Index81.6#1 of 42Adaptive reasoning, max effort
GPQA Diamond93.7%#7 of 44Adaptive reasoning, max effort
Humanity’s Last Exam59.1%#2 of 51Adaptive reasoning, max effort
AA-LCR85.3%#4 of 51Adaptive reasoning, max effort
Terminal-Bench 2.191.4%#1 of 42Adaptive reasoning, max effort
τ²-Bench (banking)47.2%#6 of 41Adaptive reasoning, max effort
SciCode63.1%#2 of 46Adaptive reasoning, max effort
BLIND HUMAN VOTES · LMARENA
LMArena Text5,783 votes · ±81499#5 of 42#5 of 402 on LMArena · 13 Sep 2026
LMArena WebDev4,887 votes · ±111755#2 of 45#2 of 129 on LMArena · 22 Sep 2026
LMArena Vision2,701 votes · ±121289#9 of 33#11 of 152 on LMArena · 13 Sep 2026
LMArena Document1,403 votes · ±151493#4 of 26#6 of 44 on LMArena · 13 Sep 2026
LMArena Agent13,320 sessions#1#1 of 37#1 of 46 on LMArena · 15 Sep 2026
§ 03 — COST

What Claude Fable 5.1 costs in Australian dollars.

List prices converted at the snapshot’s RBA rate, excluding GST. Reasoning models are billed for their thinking as output tokens, so heavier settings cost more than the job estimates show.

LIST PRICE
Claude Fable 5.1 price per million tokens
PER MILLION TOKENSAUDUSD LIST
Input tokensA$14.04US$10.00
Output tokensA$70.19US$50.00
Blended (3 in : 1 out)A$28.08US$20.00
EVERYDAY JOBSESTIMATES
Estimated cost of common jobs in Australian dollars
JOBCOST (AUD)MEDIAN MODEL
Customer support replyPER 1,000 REPLIESA$49.14A$6.58
Summarise a 30-page documentPER 100 DOCUMENTSA$33.69A$4.74
Agentic coding taskPER 10 TASKSA$26.67A$3.72
SETTINGSMEASURED SEPARATELY
Claude Fable 5.1 settings compared
SETTINGINTELLIGENCEA$ / 1MSPEED
Adaptive reasoning, max effort53.4A$28.0865 tok/s
Adaptive reasoning, xhigh effort53.2A$28.0859 tok/s
Adaptive reasoning, high effort51.2A$28.0849 tok/s
Adaptive reasoning, medium effort48.9A$28.0850 tok/s
Adaptive reasoning, low effort46.8A$28.0849 tok/s
§ 04 — AUSTRALIA

Running Claude Fable 5.1 in Australia.

It can be called from AWS Bedrock in Sydney and Melbourne and Google Vertex AI in Sydney, but only through global routing, so requests may be processed outside Australia. How the platforms compare →

AUSTRALIAN CLOUD REGIONS
Claude Fable 5.1 availability in Australian cloud regions
PLATFORMAVAILABILITYDETAIL
AWS Bedrock · SydneyGLOBAL

Provider docs, checked 23 Sep 2026

AWS Bedrock · MelbourneGLOBAL

Provider docs, checked 23 Sep 2026

Azure · Australia EastNOT IN AU

Provider docs, checked 23 Sep 2026

Google Vertex AI · SydneyGLOBAL

Provider docs, checked 23 Sep 2026

§ 06 — QUESTIONS

Claude Fable 5.1, answered.

Claude Fable 5.1 lists at US$10.00 per million input tokens and US$50.00 per million output tokens — A$14.04 and A$70.19 at A$1 = US$0.7123 (RBA, 22 Sep 2026), excluding GST. A typical customer support reply works out at about A$49.14 per 1,000 replies, before any reasoning tokens.

Claude Fable 5.1 can be called from AWS Bedrock in Sydney and Melbourne and Google Vertex AI in Sydney, but only through global routing, so requests may be processed outside Australia. Availability comes from the cloud providers’ own documentation; check your provider’s current terms before relying on it for data-residency obligations.

It ranks 1st of 42 on our coding ranking, with an Artificial Analysis Coding Index of 81.6 and an LMArena WebDev rating of 1755 (2nd of 129).

It ranks 1st of 41 on our agents ranking, completing 47.2% of τ²-Bench banking customer-service tasks and 91.4% of Terminal-Bench 2.1 tasks.

Artificial Analysis measures Claude Fable 5.1 at about 65 tok/s of output, with the answer starting after 2.0 min on average once thinking time is included (adaptive reasoning, max effort setting).

Anthropic documents a context window of 1,000,000 tokens (1M), with up to 128,000 output tokens. On AA-LCR, which tests reasoning across ~100,000-token document sets, it scores 85.3%.

PUT THE COMPARISON TO WORK

Need help choosing and using AI for your business?