RankingClaude Fable 5.1

Claude Fable 5.1

Data updated 12 Sept 2026

40 published benchmark measures · 16 benchmark families contribute across 5 task areas. 4 capability estimates available in the full profile. See all results ↓ · Compare published benchmarks →

Model evidence summary

Choose a configuration

A model-wide score would mix effort settings. This page summarizes published evidence; only identified configurations receive leaderboard ranks.

Compare measured configurations →

The capability chart summarizes evidence across reported settings, not a runnable configuration or a leaderboard rank.

Performance profile

Capabilities

Adjusted comparison score · 0–100

Filled points are supported; hollow points are preliminary. Lines stop at unknown capabilities. Exact values and sources follow below.

Agentic4 families · 0 with independent evidence · Supported97.8

4 core families; 4 direct opponents across 2 labs.

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: 97.1–98.9; 2/11 scenarios unsupported. Without one publisher: 96.8–98.7; 1/15 unsupported. Smoothing check: 92.0–99.2; 0/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: family and publisher holdouts improved on neutral predictions within evaluated coverage; source sensitivity still requires inspection. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

AutomationBench: 87.5 observed win share
Anthropic, OpenAI · Source 1 Source 2
GDPval: 100.0 observed win share
Anthropic · Source 1
OSWorld: 100.0 observed win share
Anthropic · Source 1 Source 2
OfficeQA: 100.0 observed win share
Anthropic · Source 1 Source 2

Hard reasoning5 families · 2 with independent evidence · Supported86.8

5 core families; 57 direct opponents across 12 labs.

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: 80.7–90.7; 0/7 scenarios unsupported. Without one publisher: 69.6–88.7; 0/16 unsupported. Smoothing check: 79.8–89.6; 0/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: family and publisher holdouts improved on neutral predictions within evaluated coverage. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

GPQA: 51.4 observed win share
OpenAI, artificialanalysis.ai · Source 1 Source 2
ARC-AGI: 53.4 observed win share
Anthropic, OpenAI, arcprize.org · Source 1 Source 2 Source 3 Source 4 Source 5
Humanity’s Last Exam: 100.0 observed win share
Anthropic, OpenAI · Source 1 Source 2 Source 3
FrontierMath: 50.0 observed win share
OpenAI · Source 1
Terminal-Bench Science: 75.0 observed win share
OpenAI · Source 1

Coding4 families · 0 with independent evidence · Supported89.9

4 core families; 5 direct opponents across 3 labs.

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: 87.2–93.4; 0/7 scenarios unsupported. Without one publisher: 74.1–91.8; 0/20 unsupported. Smoothing check: 82.6–93.0; 0/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: family and publisher holdouts improved on neutral predictions within evaluated coverage. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

SWE-bench Multilingual: 50.0 observed win share
Anthropic · Source 1
SWE-bench Pro: 100.0 observed win share
Anthropic · Source 1
Terminal-Bench: 90.0 observed win share
Anthropic, OpenAI · Source 1 Source 2 Source 3
DeepSWE: 0.0 observed win share
OpenAI · Source 1

Human prefNo comparable evidenceUnknown

0 core families; 0 direct opponents across 0 labs. No admitted core comparisons

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: No supported estimate; 1/1 scenarios unsupported. Without one publisher: No supported estimate; 1/1 unsupported. Smoothing check: No supported estimate; 3/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: cross-family and cross-publisher predictions could not be evaluated from the available comparison network. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

Knowledge2 families · 0 with independent evidence · PreliminaryUnknown

2 core families; 2 direct opponents across 1 labs. No connected comparison to the complete reference panel

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: No supported estimate; 4/4 scenarios unsupported. Without one publisher: No supported estimate; 5/5 unsupported. Smoothing check: No supported estimate; 3/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: almost no publisher-held-out comparisons remained estimable. Cross-source validity is unresolved. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

Gmmlu: 100.0 observed win share
Anthropic · Source 1
Milu: 100.0 observed win share
Anthropic · Source 1

Multimodal1 families · 0 with independent evidence · Preliminary95.1

1 core families; 2 direct opponents across 1 labs. Needs broader benchmark and opponent support

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: No supported estimate; 8/8 scenarios unsupported. Without one publisher: No supported estimate; 8/8 unsupported. Smoothing check: No supported estimate; 3/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: limited publisher-held-out coverage. Cross-source validity is unresolved. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

Chartography: 100.0 observed win share
Anthropic · Source 1 Source 2

Long contextNo comparable evidenceUnknown

0 core families; 0 direct opponents across 0 labs. No admitted core comparisons

Estimated against the same complete panel: gpt-5.6-sol claude-opus-4-8 gemini-3.1-pro . Indirect comparisons assume performance can be summarized by one strength within this capability.

Refit without one family: No supported estimate; 5/5 scenarios unsupported. Without one publisher: No supported estimate; 5/5 unsupported. Smoothing check: No supported estimate; 3/3 unsupported. These are sensitivity checks, not confidence intervals.

Audit: family and publisher holdouts underperformed neutral predictions. Treat this as an exploratory estimate. Validation audit.

Matched family win shares below describe source evidence; they are not averaged to produce the adjusted score.

Scores estimate outcomes against a shared reference panel; they are not accuracy percentages. Sparse or disconnected evidence cannot qualify an overall profile. Open a capability to inspect its evidence.

Compare 5 effort levels across 63 benchmark/harness combinations →

Reported effort · Mixed settings

Settings reported in this model's published benchmark results, including results outside the aggregate. Effort names are provider-specific. These are not API defaults or equal compute budgets.

Auto thinking: 2 observations
High: 1 observations
Max: 25 observations
Medium: 3 observations
XHigh: 4 observations
Not specified: 33 observations

Mixed settings means multiple settings occur in the evidence. Best across efforts means the source selected its best reported result across settings; it does not mean Max. Unspecified settings stay unknown. This model-summary chart combines reported settings. The leaderboard keeps identified configurations separate and excludes unknown effort.

Inspect each result and its source ↓ · Download effort evidence

Compare capability profiles →

Score contributions and missing evidence

16 contributing families across 5 capabilities. Fixed reference panels do not change when the catalog expands.

Capability is fitted jointly across families. Capability estimates below describe different task areas; their weighted sum is not the Capability score.

Results without reviewed compatibility or a reference match remain in the raw evidence below. Coverage counts only contributing results.

Model information & shareable badge

Lab: Anthropic
Catalog status: active
Availability: Public provider catalog; account and region restrictions may apply
Family: Claude Fable
Released: —
Context: 1,000,000 tokens
License: proprietary
Model card: https://platform.claude.com/docs/en/about-claude/model-deprecations
Default Capability family coverage: /badge/claude-fable-5-1.svg

Benchmark scores & sources

Original results, evaluation harnesses, and evidence behind this model.

Benchmark	Bucket	Score	Harness	Evidence	Source-recorded date
AA-Briefcase	Supporting evidence	1694 Reported settings & source Artificial Analysis long-horizon knowledge projects; rubric and panel pairwise judging; Claude max effort. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · Table8.1.A;8.15.4 · reviewed 2026-09-12	Published configuration Effort: Max	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
AA-Briefcase	Supporting evidence	1686 Reported settings & source Artificial Analysis; xhigh effort. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.15.4 · reviewed 2026-09-12	Published configuration Effort: XHigh	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
AA-Briefcase	Supporting evidence	1611 Reported settings & source Artificial Analysis; high effort. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.15.4 · reviewed 2026-09-12	Published configuration Effort: High	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
AA-Briefcase	Supporting evidence	61.5% Reported settings & source Artificial Analysis; max effort; component rubric pass. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.15.4 · reviewed 2026-09-12	Published configuration Effort: Max	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
AA-Briefcase	Supporting evidence	2025 rating Reported settings & source Artificial Analysis; max effort; component analytical quality. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.15.4 · reviewed 2026-09-12	Published configuration Effort: Max	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
AA-Briefcase	Supporting evidence	1495 rating Reported settings & source Artificial Analysis; max effort; component presentation. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.15.4 · reviewed 2026-09-12	Published configuration Effort: Max	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
ARC-AGI · 1	Hard reasoning	97.5% Reported settings & source ARC Prize semi-private validation; verified Fable5.1 max effort; comparator figures as reported in summary. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · Table8.1.A;8.16 · reviewed 2026-09-12	Published configuration Effort: Max contributes to capability	lab self-report	Reviewed 2026-09-12
ARC-AGI · 2	Hard reasoning	90% Reported settings & source ARC Prize semi-private validation; verified Fable5.1 max effort; comparator figures as reported in summary. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · Table8.1.A;8.16 · reviewed 2026-09-12	Published configuration Effort: Max contributes to capability	lab self-report	Reviewed 2026-09-12
ARC-AGI-1 · 1	Hard reasoning	97.5% Reported settings & source Maximum reported across reasoning efforts; OpenAI research environment or API Provider-published result; comparator measurements are not automatically independently reproduced. GPT-6 Astra: A new generation of intelligence · Abstract Reasoning table / ARC-AGI-1 / Claude Fable 5.1 · reviewed 2026-09-06	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-06
ARC-AGI-2	Hard reasoning	90%	ARC Prize verified Effort: Not specified contributes to capability	official board	2026-09-01
ARC-AGI-2 · 2	Hard reasoning	90% Reported settings & source Maximum reported across reasoning efforts; OpenAI research environment or API Provider-published result; comparator measurements are not automatically independently reproduced. GPT-6 Astra: A new generation of intelligence · Abstract Reasoning table / ARC-AGI-2 / Claude Fable 5.1 · reviewed 2026-09-06	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-06
Artificial Analysis Intelligence Index v4.1.1 · v4.1.1	Supporting evidence	65.7 index score Reported settings & source Maximum reported across reasoning efforts; OpenAI research environment or API Provider-published result; comparator measurements are not automatically independently reproduced. GPT-6 Astra: A new generation of intelligence · Professional table / Artificial Analysis Intelligence Index v4.1.1 / Claude Fable 5.1 · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-06
AutomationBench	Agentic	31.4% Reported settings & source Private held-out board; simulated business-workflow app APIs; all assertions must pass; max effort stated for Fable5.1/Opus5. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · Table8.1.A;8.15.6 · reviewed 2026-09-12	Published configuration Effort: Max contributes to capability	lab self-report	Reviewed 2026-09-12
AutomationBench · not specified	Agentic	31.4% Reported settings & source Maximum reported across reasoning efforts; OpenAI research environment or API Provider-published result; comparator measurements are not automatically independently reproduced. GPT-6 Astra: A new generation of intelligence · Professional table / AutomationBench / Claude Fable 5.1 · reviewed 2026-09-06	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-06
BenchCAD · not specified	Supporting evidence	84.3% Reported settings & source Maximum reported across reasoning efforts; OpenAI research environment or API; tools enabled; three Anthropic evaluation modifications Provider-published result; comparator measurements are not automatically independently reproduced. Astra launch footnote5: Claude scores use three modifications described in the Fable5.1 system card; not same controlled setting. GPT-6 Astra: A new generation of intelligence · Professional table / BenchCAD / Claude Fable 5.1 · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-06
BenchCAD · Vision2Code 1000-file subset	Supporting evidence	0.437 voxel IoU Reported settings & source Random1000 of17900files; five runs; adaptive thinking max; no tools; corrected camera prompt, raw shapes accepted, last code fence parsed. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.14.2 · reviewed 2026-09-12	Published configuration Effort: Max	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
BenchCAD · Vision2Code 1000-file subset	Supporting evidence	0.843 voxel IoU Reported settings & source Random1000 of17900files; five runs; adaptive thinking max; with tools; corrected camera prompt, raw shapes accepted, last code fence parsed. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.14.2 · reviewed 2026-09-12	Published configuration Effort: Max	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
Chartography	Multimodal	42.6% Reported settings & source 100tasks; adaptive thinking max; five runs; no tools; tools condition has container, standard libraries and crop tool. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.14.1 · reviewed 2026-09-12	Published configuration Effort: Max contributes to capability	lab self-report	Reviewed 2026-09-12
Chartography	Multimodal	86.2% Reported settings & source 100tasks; adaptive thinking max; five runs; with tools; tools condition has container, standard libraries and crop tool. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.14.1 · reviewed 2026-09-12	Published configuration Effort: Max contributes to capability	lab self-report	Reviewed 2026-09-12
CursorBench · 3.2.0	Supporting evidence	73.4% Reported settings & source Cursor production agent harness; independently measured by Cursor; max effort. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.8 · reviewed 2026-09-12	Published configuration Effort: Max	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
CursorBench · 3.2.0	Supporting evidence	68% Reported settings & source Cursor production agent harness; medium effort; $3.53/task. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.8 · reviewed 2026-09-12	Published configuration Effort: Medium	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
DeepSWE · 1.1	Coding	67.4% Reported settings & source Anthropic reported configuration; adaptive thinking at max effort, default sampling, five-trial mean unless section specifies otherwise; context at most 1M. Comparator configuration follows cited prior card or board. 113 tasks; original hidden-test grading. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.3 · reviewed 2026-09-12	Published configuration Effort: Max	lab self-report No matched opponent in this evaluation unit	Reviewed 2026-09-12
DeepSWE v1.1	Coding	67.4%	Anthropic DeepSWE1.1 reported system Effort: Not specified	lab self-report Legacy self-report lacks a reviewed comparison configuration	2026-09-01
DeepSWE v1.1 · v1.1	Coding	67.4% Reported settings & source Maximum reported across reasoning efforts; OpenAI research environment or API Provider-published result; comparator measurements are not automatically independently reproduced. GPT-6 Astra: A new generation of intelligence · Coding table / DeepSWE v1.1 / Claude Fable 5.1 · reviewed 2026-09-06	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-06
FrontierCode · 1.1 Extended	Supporting evidence	63.6% Reported settings & source Cognition agentic coding; composite functional and code-quality score. Fable5.1 medium effort; Fable5 xhigh. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.4 · reviewed 2026-09-12	Published configuration Effort: Medium	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
FrontierCode · 1.1 Main	Supporting evidence	50.9% Reported settings & source Cognition agentic coding; composite functional and code-quality score. Fable5.1 medium effort; Fable5 xhigh. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.4 · reviewed 2026-09-12	Published configuration Effort: Medium	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
FrontierCode 1.1 Extended (score) · 1.1	Supporting evidence	63.6% Reported settings & source Maximum reported across reasoning efforts; OpenAI research environment or API Provider-published result; comparator measurements are not automatically independently reproduced. GPT-6 Astra: A new generation of intelligence · Coding table / FrontierCode 1.1 Extended (score) / Claude Fable 5.1 · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-06
FrontierCode 1.1 Main (score) · 1.1	Supporting evidence	50.9% Reported settings & source Maximum reported across reasoning efforts; OpenAI research environment or API Provider-published result; comparator measurements are not automatically independently reproduced. GPT-6 Astra: A new generation of intelligence · Coding table / FrontierCode 1.1 Main (score) / Claude Fable 5.1 · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-06
FrontierMath Tier 4 (v2) · v2	Hard reasoning	87.8% Reported settings & source Maximum reported across reasoning efforts; OpenAI research environment or API Provider-published result; comparator measurements are not automatically independently reproduced. GPT-6 Astra: A new generation of intelligence · Academic table / FrontierMath Tier 4 (v2) / Claude Fable 5.1 · reviewed 2026-09-06	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-06
FrontierSWE · 2	Supporting evidence	0.57 fraction Reported settings & source Proximal agent harness; max effort; 34 tasks, five trials/task; mean score on 0..1 scale. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.5 · reviewed 2026-09-12	Published configuration Effort: Max	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
GDPval-AA · 2	Agentic	1853 Reported settings & source Artificial Analysis independent agentic shell/web evaluation;220GDPvalgoldtasks; blind pairwise Elo; max effort for Claude. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · Table8.1.A;8.15.3 · reviewed 2026-09-12	Published configuration Effort: Max contributes to capability	lab self-report	Reviewed 2026-09-12
GDPval-AA · 2	Agentic	1835 Reported settings & source Artificial Analysis; xhigh effort; same release board snapshot. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.15.3 · reviewed 2026-09-12	Published configuration Effort: XHigh	lab self-report No matched opponent in this evaluation unit	Reviewed 2026-09-12
GMMLU	Knowledge	94% Reported settings & source Mean accuracy42languages; adaptive max; one trial; no tools/custom system prompts. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.18.1 · reviewed 2026-09-12	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-12
GPQA Diamond	Hard reasoning	93.737%	GPQA Diamond reported Effort: Not specified contributes to capability	independent repro	2026-09-11
GPQA Diamond · not specified	Hard reasoning	93.7% Reported settings & source Maximum reported across reasoning efforts; OpenAI research environment or API Provider-published result; comparator measurements are not automatically independently reproduced. GPT-6 Astra: A new generation of intelligence · Academic table / GPQA Diamond / Claude Fable 5.1 · reviewed 2026-09-06	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-06
HealthBench	Supporting evidence	66.7% Reported settings & source Raw rubric score; adaptive max; five trials; no tools/custom system prompt; Opus4.8grader; Fable5.1 safety fallback toOpus5. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.17 · reviewed 2026-09-12	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
HealthBench	Supporting evidence	60% Reported settings & source Length-adjusted score using GPT5.5card method; otherwise raw evaluation configuration. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.17.1 · reviewed 2026-09-12	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
HealthBench Professional	Supporting evidence	74.2% Reported settings & source Raw rubric score; adaptive max; five trials; no tools/custom system prompt; Opus4.8grader; Fable5.1 safety fallback toOpus5. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.17 · reviewed 2026-09-12	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
HealthBench Professional	Supporting evidence	62.1% Reported settings & source Length-adjusted score; HealthBench Professional paper method; no tools; Opus4.8grader; five trials. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · Table8.1.A;8.17.2 · reviewed 2026-09-12	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
HealthBench Professional (length-adjusted) · not specified	Supporting evidence	58.1% Reported settings & source Maximum reported across reasoning efforts; OpenAI research environment or API; official paper scoring; length-adjusted; OpenAI reproduction; GPT-5.4 grader; unclipped; Opus5 fallback for provider refusals Provider-published result; comparator measurements are not automatically independently reproduced. GPT-6 Astra: A new generation of intelligence · Science And Health table / HealthBench Professional (length-adjusted) / Claude Fable 5.1 · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-06
Humanity's Last Exam	Hard reasoning	60.9%	HLE no tools Effort: Not specified	lab self-report Legacy self-report lacks a reviewed comparison configuration	2026-09-01
Humanity's Last Exam	Hard reasoning	65%	HLE with tools Effort: Not specified	lab self-report Legacy self-report lacks a reviewed comparison configuration	2026-09-01
Humanity’s Last Exam	Hard reasoning	60.9% Reported settings & source Full2500questions; no tools; auto thinking;1Mtotal token cap; no compaction; Opus4.6grader; restricted fetch and contamination review for tools. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · Table 8.1.A;8.12.1 · reviewed 2026-09-12	Published configuration Effort: Auto thinking contributes to capability	lab self-report	Reviewed 2026-09-12
Humanity’s Last Exam	Hard reasoning	65% Reported settings & source Full2500questions; with tools; auto thinking;1Mtotal token cap; no compaction; Opus4.6grader; restricted fetch and contamination review for tools. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · Table 8.1.A;8.12.1 · reviewed 2026-09-12	Published configuration Effort: Auto thinking contributes to capability	lab self-report	Reviewed 2026-09-12
Humanity's Last Exam (w/ tools) · not specified	Hard reasoning	65% Reported settings & source Maximum reported across reasoning efforts; OpenAI research environment or API; tools enabled Provider-published result; comparator measurements are not automatically independently reproduced. GPT-6 Astra: A new generation of intelligence · Academic table / Humanity's Last Exam (w/ tools) / Claude Fable 5.1 · reviewed 2026-09-06	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-06
Internal Database Migration Tasks · not specified	Supporting evidence	57.8% Reported settings & source Maximum reported across reasoning efforts; OpenAI research environment or API Provider-published result; comparator measurements are not automatically independently reproduced. GPT-6 Astra: A new generation of intelligence · Coding table / Internal Database Migration Tasks / Claude Fable 5.1 · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-06
Legal Agent Benchmark · 120-task held-out subset	Supporting evidence	16.7% Reported settings & source all-pass; Artificial Analysis harness; xhigh effort. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.15.2 · reviewed 2026-09-12	Published configuration Effort: XHigh	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
Legal Agent Benchmark · 120-task held-out subset	Supporting evidence	93.3% Reported settings & source criterion-pass; Artificial Analysis harness; xhigh effort. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.15.2 · reviewed 2026-09-12	Published configuration Effort: XHigh	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
Legal Agent Benchmark · 1235-task public subset	Supporting evidence	19.09% Reported settings & source all-pass; five runs; adaptive max; internal bash/Python harness, Sonnet4.6judge;16defective tasks excluded; production safeguards/fallback. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.15.2 · reviewed 2026-09-12	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
Legal Agent Benchmark · 1235-task public subset	Supporting evidence	90.81% Reported settings & source criterion-pass; five runs; adaptive max; internal bash/Python harness, Sonnet4.6judge;16defective tasks excluded; production safeguards/fallback. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.15.2 · reviewed 2026-09-12	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
MILU	Knowledge	93% Reported settings & source Mean accuracy11languages; adaptive max; five trials; no tools/custom system prompts. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.18.2 · reviewed 2026-09-12	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-12
OfficeQA	Agentic	80.2% Reported settings & source Extracted-text Treasury corpus in sandbox; code execution; production Messages API with safeguards/fallback;128koutput cap. Pro133questions. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.15.1 · reviewed 2026-09-12	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-12
OfficeQA Pro	Agentic	69% Reported settings & source Extracted-text Treasury corpus in sandbox; code execution; production Messages API with safeguards/fallback;128koutput cap. Pro133questions. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.15.1 · reviewed 2026-09-12	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-12
OSWorld · 2.0 August2026 task release	Agentic	77.9% Reported settings & source partial pass@1;108tasks; five runs;1080p;500steps; max effort; Opus4.8grader; task fixes; Fable safety interventions score zero. Same-condition reruns; supersedes earlier OSWorld2results and incompatible with previous task files. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.14.3 · reviewed 2026-09-12	Published configuration Effort: Max contributes to capability	lab self-report	Reviewed 2026-09-12
OSWorld · 2.0 August2026 task release	Agentic	41.7% Reported settings & source strict pass@1;108tasks; five runs;1080p;500steps; max effort; Opus4.8grader; task fixes; Fable safety interventions score zero. Same-condition reruns; supersedes earlier OSWorld2results and incompatible with previous task files. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.14.3 · reviewed 2026-09-12	Published configuration Effort: Max contributes to capability	lab self-report	Reviewed 2026-09-12
ProgramBench · 166 golden-task subset	Supporting evidence	87.6% Reported settings & source mini-swe-agent without six-hour timeout; excludes34flaky-reference tasks; tests restricted to reference-passing tests; up to1Mcontext. Hidden-test pass rate, not fraction of completely solved programs. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.11.1 · reviewed 2026-09-12	Published configuration Effort: Not specified	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
SWE-bench Multilingual	Coding	89.1% Reported settings & source Anthropic reported configuration; adaptive thinking at max effort, default sampling, five-trial mean unless section specifies otherwise; context at most 1M. Comparator configuration follows cited prior card or board. Agent identity is not established as mini-swe-agent. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · Table 8.1.A; section 8.2 · reviewed 2026-09-12	Published configuration Effort: Max contributes to capability	lab self-report	Reviewed 2026-09-12
SWE-bench Multimodal	Supporting evidence	54.7% Reported settings & source Anthropic reported configuration; adaptive thinking at max effort, default sampling, five-trial mean unless section specifies otherwise; context at most 1M. Comparator configuration follows cited prior card or board. Agent identity is not established as mini-swe-agent. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · Table 8.1.A; section 8.2 · reviewed 2026-09-12	Published configuration Effort: Max	lab self-report Supporting evidence outside the reviewed capability core	Reviewed 2026-09-12
SWE-bench Pro	Coding	81.2%	Anthropic SWE-bench Pro reported system Effort: Not specified	lab self-report Legacy self-report lacks a reviewed comparison configuration	2026-09-01
SWE-bench Pro	Coding	81.2% Reported settings & source Anthropic reported configuration; adaptive thinking at max effort, default sampling, five-trial mean unless section specifies otherwise; context at most 1M. Comparator configuration follows cited prior card or board. Agent identity is not established as mini-swe-agent. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · Table 8.1.A; section 8.2 · reviewed 2026-09-12	Published configuration Effort: Max contributes to capability	lab self-report	Reviewed 2026-09-12
Terminal-Bench · 4.0	Coding	55.8% Reported settings & source Claude Code --bare max effort, 15 trials/task over 66 tasks; Anthropic internal reruns. Uses more precise section values rather than rounded headline table. Distinct from2.1; timeouts/resources changed. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.6 · reviewed 2026-09-12	Published configuration Effort: Max contributes to capability	lab self-report	Reviewed 2026-09-12
Terminal-Bench 4.0 · 4.0	Coding	55.8% Reported settings & source Maximum reported across reasoning efforts; OpenAI research environment or API Provider-published result; comparator measurements are not automatically independently reproduced. GPT-6 Astra: A new generation of intelligence · Coding table / Terminal-Bench 4.0 / Claude Fable 5.1 · reviewed 2026-09-06	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-06
Terminal-Bench Science 0.1 · not specified	Hard reasoning	52.6% Reported settings & source Maximum reported across reasoning efforts; OpenAI research environment or API Provider-published result; comparator measurements are not automatically independently reproduced. GPT-6 Astra: A new generation of intelligence · Academic table / Terminal-Bench Science 0.1 / Claude Fable 5.1 · reviewed 2026-09-06	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-06
Terminal-Bench-Science · 0.1	Coding	52.6% Reported settings & source 70tasks; Claude Code --bare max; Fable10trials/task, Opus12; GPT Codex CLI max from public board. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · 8.7 · reviewed 2026-09-12	Published configuration Effort: Not specified contributes to capability	lab self-report	Reviewed 2026-09-12
Toolathlon · Verified June2026	Agentic	77.8% Reported settings & source Pass@1;108tasks/three trials; internal harness; max effort; Fable safeguards+Opus4.8fallback; Opus safeguards/fallback disabled; pinned containers/data. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · Table8.15.5.A · reviewed 2026-09-12	Published configuration Effort: Max	lab self-report System card section8.15.5 confirms11of324trials were partly or fully completed by Opus4.8fallback; actual mixed-model execution, not merely available fallback.	Reviewed 2026-09-12
Toolathlon · Verified June2026	Agentic	81.5% Reported settings & source Pass@3;108tasks/three trials; internal harness; max effort; Fable safeguards+Opus4.8fallback; Opus safeguards/fallback disabled; pinned containers/data. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · Table8.15.5.A · reviewed 2026-09-12	Published configuration Effort: Max	lab self-report System card section8.15.5 confirms11of324trials were partly or fully completed by Opus4.8fallback; actual mixed-model execution, not merely available fallback.	Reviewed 2026-09-12
Toolathlon · Verified June2026	Agentic	73.1% Reported settings & source Pass³;108tasks/three trials; internal harness; max effort; Fable safeguards+Opus4.8fallback; Opus safeguards/fallback disabled; pinned containers/data. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · Table8.15.5.A · reviewed 2026-09-12	Published configuration Effort: Max	lab self-report System card section8.15.5 confirms11of324trials were partly or fully completed by Opus4.8fallback; actual mixed-model execution, not merely available fallback.	Reviewed 2026-09-12
Toolathlon · Verified June2026	Agentic	23.7 turns Reported settings & source average turns;108tasks/three trials; internal harness; max effort; Fable safeguards+Opus4.8fallback; Opus safeguards/fallback disabled; pinned containers/data. Fable is the safeguarded deployed configuration; selected tasks may use disclosed Opus fallback, except evaluations explicitly counting safety blocks as failures. Claude Fable 5.1 and Claude Mythos 5.1 System Card · Table8.15.5.A · reviewed 2026-09-12	Published configuration Effort: Max	lab self-report System card section8.15.5 confirms11of324trials were partly or fully completed by Opus4.8fallback; actual mixed-model execution, not merely available fallback.	Reviewed 2026-09-12
GDPval-AA	Agentic	—	—	—	—
LiveCodeBench	Coding	—	—	—	—
LMArena Text Arena	Human pref	—	—	—	—
MMLU-Pro	Knowledge	—	—	—	—
OSWorld-Verified	Agentic	—	—	—	—
SWE-bench Verified	Agentic	—	—	—	—
Terminal-Bench 2.1	Agentic	—	—	—	—