RankingGPT-6 Astra

GPT-6 Astra

Data updated 22 Oct 2026

85 published benchmark measures · 13 benchmark families contribute across 5 task areas. 5 capability estimates available in the full profile. See all results ↓ · Compare published benchmarks →

Model evidence summary

This profile combines published settings. It is not a runnable configuration or a leaderboard rank. Compare measured configurations →

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.

Agentic5 families · 0 with independent evidence · Supported98.1

5 core families; 6 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: 97.0–98.6; 0/13 scenarios unsupported. Without one publisher: 97.4–98.5; 1/15 unsupported. Smoothing check: 91.8–99.6; 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.

Browsecomp: 100.0 observed win share
OpenAI · Source 1
AutomationBench: 100.0 observed win share
OpenAI · Source 1 Source 2 Source 3 Source 4 Source 5 Source 6
OSWorld: 100.0 observed win share
OpenAI · Source 1 Source 2 Source 3 Source 4 Source 5 Source 6
Terminal-Bench Science: 100.0 observed win share
OpenAI · Source 1 Source 2 Source 3 Source 4 Source 5 Source 6
GDP.pdf: 85.0 observed win share
OpenAI · Source 1 Source 2 Source 3 Source 4 Source 5

Hard reasoning4 families · 2 with independent evidence · Supported90.3

4 core families; 61 direct opponents across 10 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: 79.7–95.9; 0/6 scenarios unsupported. Without one publisher: 86.9–91.1; 0/16 unsupported. Smoothing check: 83.5–92.7; 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: 100.0 observed win share
OpenAI, artificialanalysis.ai · Source 1 Source 2
ARC-AGI: 96.9 observed win share
OpenAI, arcprize.org · Source 1 Source 2 Source 3
FrontierMath: 100.0 observed win share
OpenAI · Source 1
Humanity’s Last Exam: 0.0 observed win share
OpenAI · Source 1

Coding2 families · 2 with independent evidence · Supported97.8

2 core families; 35 direct opponents across 9 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: 97.8–98.0; 2/6 scenarios unsupported. Without one publisher: 94.6–98.1; 0/19 unsupported. Smoothing check: 91.0–99.5; 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.

Terminal-Bench: 96.4 observed win share
OpenAI, tbench.ai · Source 1 Source 2
DeepSWE: 91.7 observed win share
OpenAI, deepswe.datacurve.ai · Source 1 Source 2 Source 3 Source 4 Source 5 Source 6 Source 7

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.

KnowledgeNo 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; 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.

Multimodal1 families · 0 with independent evidence · Preliminary94.6

1 core families; 1 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.

ScreenSpot: 100.0 observed win share
OpenAI · Source 1

Long context1 families · 0 with independent evidence · Preliminary94.4

1 core families; 1 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; 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.

MRCR: 100.0 observed win share
OpenAI · Source 1 Source 2

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.

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.

Best across efforts: 34 observations
High: 6 observations
Low: 6 observations
Max: 10 observations
Medium: 6 observations
XHigh: 7 observations
Not specified: 62 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

Score contributions and missing evidence

13 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: OpenAI
Catalog status: active
Availability: Public provider catalog; account and region restrictions may apply
Family: GPT-6
Released: —
Context: 1,050,000 tokens
License: proprietary
Model card: https://developers.openai.com/api/docs/models/all
Default Capability family coverage: /badge/gpt-6-astra.svg

Benchmark scores & sources

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

Benchmark	Bucket	Score	Harness	Evidence	Source-recorded date
FrontierSWE · 2	Supporting evidence	65.5% Reported settings & source Comparator maximum available thinking/reasoning when reported, otherwise best available result; exact setting not identified by this grid. Results sourced from Proximal's official public leaderboard. Provider-published launch claim, not independent measurement by Google for every row. Results sourced from Proximal's official public leaderboard. Methodology https://deepmind.google/models/evals-methodology/gemini-4-argon, SHA256 ff1df6bdeddc4c0f48840c09a8a7813b10ae7ad9e053892da68dabf64f09bff7. Preserve source rounding. Methodology says capabilities as of September 2026 and results as of October 2026; review date is source retrieval, not evaluation date. Comparison limit: Provider-published claim. Independent source results are admitted separately with exact checkpoint, effort and compatible protocol; this grid does not establish a matched comparison. Historical provider claim; secondary and superseded for current FrontierSWE comparisons by the independent Proximal FrontierSWE v2 mean@5 board at https://www.frontierswe.com/. Version-unspecified claims are not v2 results. Excluded from weighted Overall. Google Gemini 4 Argon launch and evaluation methodology · Launch grid row 6: FrontierSWE 2 / GPT-6 Astra; methodology Additional Details · reviewed 2026-09-30	Published configuration Effort: Not specified	lab self-report Provider-published claim. Independent source results are admitted separately with exact checkpoint, effort and compatible protocol; this grid does not establish a matched comparison. Historical provider claim; secondary and superseded for current FrontierSWE comparisons by the independent Proximal FrontierSWE v2 mean@5 board at https://www.frontierswe.com/. Version-unspecified claims are not v2 results. Excluded from weighted Overall.	Reviewed 2026-09-30
FrontierSWE v2	Supporting evidence	65.5%	Proximal FrontierSWE v2 · Proximus · unspecified Effort: Not specified	official board Supporting evidence outside the reviewed capability core	2026-10-22
gdp.pdf · not specified	Agentic	30.4% Reported settings & source Reported reasoning effort low. Professional questions about complex PDFs across ten professional domains. OpenAI research environment or API; same named metric and evaluation chart. No fallback is reported for this OpenAI configuration. Provider-published lab self-report. Published chart cost per task $1.6963186; source-specific cost does not establish consensus workload efficiency. Competitor results are republished from public reports; the page does not establish independent reproduction by OpenAI. Introducing GPT-6.1 Sol · Chart: GDP.pdf / GPT-6 Astra / low · reviewed 2026-09-29	Published configuration Effort: Low contributes to capability	lab self-report	Reviewed 2026-09-29
gdp.pdf · not specified	Agentic	30.4% Reported settings & source Reported reasoning effort medium. Professional questions about complex PDFs across ten professional domains. OpenAI research environment or API; same named metric and evaluation chart. No fallback is reported for this OpenAI configuration. Provider-published lab self-report. Published chart cost per task $1.7198086; source-specific cost does not establish consensus workload efficiency. Competitor results are republished from public reports; the page does not establish independent reproduction by OpenAI. Introducing GPT-6.1 Sol · Chart: GDP.pdf / GPT-6 Astra / medium · reviewed 2026-09-29	Published configuration Effort: Medium contributes to capability	lab self-report	Reviewed 2026-09-29
gdp.pdf · not specified	Agentic	31% Reported settings & source Reported reasoning effort high. Professional questions about complex PDFs across ten professional domains. OpenAI research environment or API; same named metric and evaluation chart. No fallback is reported for this OpenAI configuration. Provider-published lab self-report. Published chart cost per task $1.7948240999999998; source-specific cost does not establish consensus workload efficiency. Competitor results are republished from public reports; the page does not establish independent reproduction by OpenAI. Introducing GPT-6.1 Sol · Chart: GDP.pdf / GPT-6 Astra / high · reviewed 2026-09-29	Published configuration Effort: High contributes to capability	lab self-report	Reviewed 2026-09-29
gdp.pdf · not specified	Agentic	32.2% Reported settings & source Reported reasoning effort xhigh. Professional questions about complex PDFs across ten professional domains. OpenAI research environment or API; same named metric and evaluation chart. No fallback is reported for this OpenAI configuration. Provider-published lab self-report. Published chart cost per task $1.9127230999999998; source-specific cost does not establish consensus workload efficiency. Competitor results are republished from public reports; the page does not establish independent reproduction by OpenAI. Introducing GPT-6.1 Sol · Chart: GDP.pdf / GPT-6 Astra / xhigh · reviewed 2026-09-29	Published configuration Effort: XHigh contributes to capability	lab self-report	Reviewed 2026-09-29
gdp.pdf · not specified	Agentic	31% Reported settings & source Reported reasoning effort max. Professional questions about complex PDFs across ten professional domains. OpenAI research environment or API; same named metric and evaluation chart. No fallback is reported for this OpenAI configuration. Provider-published lab self-report. Published chart cost per task $2.0758207; source-specific cost does not establish consensus workload efficiency. Competitor results are republished from public reports; the page does not establish independent reproduction by OpenAI. Introducing GPT-6.1 Sol · Chart: GDP.pdf / GPT-6 Astra / max · reviewed 2026-09-29	Published configuration Effort: Max contributes to capability	lab self-report	Reviewed 2026-09-29
GDPval-AA · 2	Agentic	1542 Reported settings & source Grok 4.7 XHigh; Grok 4.6 High; Claude Fable 5.1 Max; GPT-6 Astra Max. GDPval launch chart. Launch chart rounded Elo. Original evaluator: https://artificialanalysis.ai/evaluations/gdpval-aa Comparison limit: Secondary Artificial Analysis leaderboard report; retain the original evaluator once, without treating the launch reprint as additional evidence. Grok 4.7 launch comparison · Professional knowledge work: GDPval chart · reviewed 2026-09-21	Published configuration Effort: Max	lab self-report Secondary Artificial Analysis leaderboard report; retain the original evaluator once, without treating the launch reprint as additional evidence.	Reviewed 2026-09-21
GeneBench Pro · v13	Supporting evidence	37.1% 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 · Science And Health table / GeneBench Pro / GPT‑6 Astra · reviewed 2026-09-06	Published configuration Effort: Best across efforts	lab self-report Narrow specialist biological, medical or security evaluation; preserved as raw evidence outside general capability scoring.	Reviewed 2026-09-06
GPQA Diamond	Hard reasoning	96.061%	GPQA Diamond reported Effort: Not specified contributes to capability	independent repro	2026-10-20
GPQA Diamond · not specified	Hard reasoning	96% 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 / GPT‑6 Astra · reviewed 2026-09-06	Published configuration Effort: Best across efforts contributes to capability	lab self-report	Reviewed 2026-09-06
GPQA Diamond · not specified	Hard reasoning	94.9% Reported settings & source Lower-cost setting; exact effort not specified Provider-published result; preserves source setting without claiming a controlled cross-provider comparison. GPT-6 Astra: A new generation of intelligence · GPQA Diamond chart caption · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report No matched opponent in this evaluation unit	Reviewed 2026-09-06
GraphWalks · 256K to 1M BFS F1	Supporting evidence	71.8% Reported settings & source Comparator maximum available thinking/reasoning when reported, otherwise best available result; exact setting not identified by this grid. All models self-computed. 200 problems with context length between 256K and 1M tokens. BFS F1, not pass rate. Provider-published launch claim, not independent measurement by Google for every row. All models self-computed. 200 problems with context length between 256K and 1M tokens. BFS F1, not pass rate. Methodology https://deepmind.google/models/evals-methodology/gemini-4-argon, SHA256 ff1df6bdeddc4c0f48840c09a8a7813b10ae7ad9e053892da68dabf64f09bff7. Preserve source rounding. Methodology says capabilities as of September 2026 and results as of October 2026; review date is source retrieval, not evaluation date. Comparison limit: Provider-published claim. Independent source results are admitted separately with exact checkpoint, effort and compatible protocol; this grid does not establish a matched comparison. Google Gemini 4 Argon launch and evaluation methodology · Launch grid row 14: GraphWalks 256K to 1M BFS F1 / GPT-6 Astra; methodology Additional Details · reviewed 2026-09-30	Published configuration Effort: Not specified	lab self-report Provider-published claim. Independent source results are admitted separately with exact checkpoint, effort and compatible protocol; this grid does not establish a matched comparison.	Reviewed 2026-09-30
GraphWalks · up to 128K BFS F1	Supporting evidence	98.7% Reported settings & source Comparator maximum available thinking/reasoning when reported, otherwise best available result; exact setting not identified by this grid. All models self-computed. 650 items with context length up to 128K tokens. BFS F1, not pass rate. Provider-published launch claim, not independent measurement by Google for every row. All models self-computed. 650 items with context length up to 128K tokens. BFS F1, not pass rate. Methodology https://deepmind.google/models/evals-methodology/gemini-4-argon, SHA256 ff1df6bdeddc4c0f48840c09a8a7813b10ae7ad9e053892da68dabf64f09bff7. Preserve source rounding. Methodology says capabilities as of September 2026 and results as of October 2026; review date is source retrieval, not evaluation date. Comparison limit: Provider-published claim. Independent source results are admitted separately with exact checkpoint, effort and compatible protocol; this grid does not establish a matched comparison. Google Gemini 4 Argon launch and evaluation methodology · Launch grid row 13: GraphWalks up to 128K BFS F1 / GPT-6 Astra; methodology Additional Details · reviewed 2026-09-30	Published configuration Effort: Not specified	lab self-report Provider-published claim. Independent source results are admitted separately with exact checkpoint, effort and compatible protocol; this grid does not establish a matched comparison.	Reviewed 2026-09-30
Harvey's Legal Agent Benchmark · 1	Supporting evidence	5.4% Reported settings & source Comparator maximum available thinking/reasoning when reported, otherwise best available result; exact setting not identified by this grid. Results sourced from Vals AI. Provider-published launch claim, not independent measurement by Google for every row. Results sourced from Vals AI. Methodology https://deepmind.google/models/evals-methodology/gemini-4-argon, SHA256 ff1df6bdeddc4c0f48840c09a8a7813b10ae7ad9e053892da68dabf64f09bff7. Preserve source rounding. Methodology says capabilities as of September 2026 and results as of October 2026; review date is source retrieval, not evaluation date. Comparison limit: Provider-published claim. Independent source results are admitted separately with exact checkpoint, effort and compatible protocol; this grid does not establish a matched comparison. Google Gemini 4 Argon launch and evaluation methodology · Launch grid row 4: Harvey's Legal Agent Benchmark 1 / GPT-6 Astra; methodology Additional Details · reviewed 2026-09-30	Published configuration Effort: Not specified	lab self-report Provider-published claim. Independent source results are admitted separately with exact checkpoint, effort and compatible protocol; this grid does not establish a matched comparison.	Reviewed 2026-09-30
HealthBench · not specified	Supporting evidence	58.1 score (0-100) Reported settings & source length-adjusted; official HealthBench scoring Mean response length 2258 characters. Adjustment centered on2000characters; adjusted and unadjusted values remain distinct configurations. GPT-6 Astra System Card · Table6 / HealthBench / gpt-6-astra / length-adjusted · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Narrow specialist biological, medical or security evaluation; preserved as raw evidence outside general capability scoring.	Reviewed 2026-09-06
HealthBench · not specified	Supporting evidence	59.7 score (0-100) Reported settings & source unadjusted; official HealthBench scoring Mean response length 2258 characters. Adjustment centered on2000characters; adjusted and unadjusted values remain distinct configurations. GPT-6 Astra System Card · Table6 / HealthBench / gpt-6-astra / unadjusted · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Narrow specialist biological, medical or security evaluation; preserved as raw evidence outside general capability scoring.	Reviewed 2026-09-06
HealthBench Consensus · not specified	Supporting evidence	95.8 score (0-100) Reported settings & source length-adjusted; official HealthBench scoring Mean response length 2237 characters. Adjustment centered on2000characters; adjusted and unadjusted values remain distinct configurations. GPT-6 Astra System Card · Table6 / HealthBench Consensus / gpt-6-astra / length-adjusted · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Narrow specialist biological, medical or security evaluation; preserved as raw evidence outside general capability scoring.	Reviewed 2026-09-06
HealthBench Consensus · not specified	Supporting evidence	95.9 score (0-100) Reported settings & source unadjusted; official HealthBench scoring Mean response length 2237 characters. Adjustment centered on2000characters; adjusted and unadjusted values remain distinct configurations. GPT-6 Astra System Card · Table6 / HealthBench Consensus / gpt-6-astra / unadjusted · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Narrow specialist biological, medical or security evaluation; preserved as raw evidence outside general capability scoring.	Reviewed 2026-09-06
HealthBench Hard · not specified	Supporting evidence	36.3 score (0-100) Reported settings & source length-adjusted; official HealthBench scoring Mean response length 2192 characters. Adjustment centered on2000characters; adjusted and unadjusted values remain distinct configurations. GPT-6 Astra System Card · Table6 / HealthBench Hard / gpt-6-astra / length-adjusted · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Narrow specialist biological, medical or security evaluation; preserved as raw evidence outside general capability scoring.	Reviewed 2026-09-06
HealthBench Hard · not specified	Supporting evidence	37.8 score (0-100) Reported settings & source unadjusted; official HealthBench scoring Mean response length 2192 characters. Adjustment centered on2000characters; adjusted and unadjusted values remain distinct configurations. GPT-6 Astra System Card · Table6 / HealthBench Hard / gpt-6-astra / unadjusted · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Narrow specialist biological, medical or security evaluation; preserved as raw evidence outside general capability scoring.	Reviewed 2026-09-06
HealthBench Professional · not specified	Supporting evidence	63.4 score (0-100) Reported settings & source length-adjusted; official HealthBench scoring Mean response length 4097 characters. Adjustment centered on2000characters; adjusted and unadjusted values remain distinct configurations. GPT-6 Astra System Card · Table6 / HealthBench Professional / gpt-6-astra / length-adjusted · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Narrow specialist biological, medical or security evaluation; preserved as raw evidence outside general capability scoring.	Reviewed 2026-09-06
HealthBench Professional · not specified	Supporting evidence	69.5 score (0-100) Reported settings & source unadjusted; official HealthBench scoring Mean response length 4097 characters. Adjustment centered on2000characters; adjusted and unadjusted values remain distinct configurations. GPT-6 Astra System Card · Table6 / HealthBench Professional / gpt-6-astra / unadjusted · reviewed 2026-09-06	Published configuration Effort: Not specified	lab self-report Narrow specialist biological, medical or security evaluation; preserved as raw evidence outside general capability scoring.	Reviewed 2026-09-06
HealthBench Professional (length-adjusted) · not specified	Supporting evidence	63.4% Reported settings & source Maximum reported across reasoning efforts; OpenAI research environment or API; official paper scoring; length-adjusted 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) / GPT‑6 Astra · reviewed 2026-09-06	Published configuration Effort: Best across efforts	lab self-report Narrow specialist biological, medical or security evaluation; preserved as raw evidence outside general capability scoring.	Reviewed 2026-09-06
Humanity's Last Exam (w/ tools) · not specified	Hard reasoning	57.2% 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) / GPT‑6 Astra · reviewed 2026-09-06	Published configuration Effort: Best across efforts contributes to capability	lab self-report	Reviewed 2026-09-06