RankingClaude Opus 5.5

Claude Opus 5.5

Data updated 31 Oct 2026

51 published benchmark measures · 1 benchmark families contribute across 1 task areas. 1 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.

AgenticNo 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; 13/13 scenarios unsupported. Without one publisher: No supported estimate; 15/15 unsupported. Smoothing check: No supported estimate; 3/3 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.

Hard reasoning1 families · 1 with independent evidence · Preliminary83.8

1 core families; 59 direct opponents across 10 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; 6/6 scenarios unsupported. Without one publisher: No supported estimate; 16/16 unsupported. Smoothing check: No supported estimate; 3/3 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.

ARC-AGI: 96.6 observed win share
arcprize.org · Source 1

CodingNo 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; 6/6 scenarios unsupported. Without one publisher: No supported estimate; 19/19 unsupported. Smoothing check: No supported estimate; 3/3 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.

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.

MultimodalNo 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; 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.

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.

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: 4 observations
Low: 2 observations
Max: 34 observations
Medium: 5 observations
XHigh: 6 observations
Not specified: 37 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

1 contributing families across 1 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 Opus
Released: 2026-09-22
Context: 1,000,000 tokens
API list price: $4 input / $20 output per million tokens
License: proprietary
Model card: https://platform.claude.com/docs/en/models/opus-5-5/overview
Default Capability family coverage: /badge/claude-opus-5-5.svg

Benchmark scores & sources

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

Benchmark	Bucket	Score	Harness	Evidence	Source-recorded date
CursorBench · 4.0	Supporting evidence	56% Reported settings & source Cursor production agent harness; high effort. Independently measured by Cursor. Provider-published lab self-report from the Claude Opus 5.5 system card (SHA256 7311c9c6bbb16d012f1c12c7418b05949fcf7ae3e30d2c40f22050074b2a7378). Not an independent board. Comparison limit: Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison. Claude Opus 5.5 System Card · section 8.8 · reviewed 2026-09-22	Published configuration Effort: High	lab self-report Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison.	Reviewed 2026-09-22
CursorBench · 4.0	Supporting evidence	52.5% Reported settings & source Cursor production agent harness; medium effort. Independently measured by Cursor. Provider-published lab self-report from the Claude Opus 5.5 system card (SHA256 7311c9c6bbb16d012f1c12c7418b05949fcf7ae3e30d2c40f22050074b2a7378). Not an independent board. Launch-page prose also states 52.5% at default (medium) effort. Comparison limit: Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison. Claude Opus 5.5 System Card · section 8.8 · reviewed 2026-09-22	Published configuration Effort: Medium	lab self-report Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison.	Reviewed 2026-09-22
CWE-bench · 1	Supporting evidence	67% Reported settings & source Comparator maximum available thinking/reasoning when reported, otherwise best available result; exact setting not identified by this grid. Official public leaderboard; ranking uses pass@1 with pass@4 tiebreaks. Grid records pass@1 only. Provider-published launch claim, not independent measurement by Google for every row. Official public leaderboard; ranking uses pass@1 with pass@4 tiebreaks. Grid records pass@1 only. 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 19: CWE-bench 1 / Claude Opus 5.5; 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
DAYJOB Finance	Supporting evidence	42.5%	DAYJOB Finance · max Effort: Not specified	official board Supporting evidence outside the reviewed capability core	2026-10-28
DAYJOB Healthcare	Supporting evidence	48.3%	DAYJOB Healthcare · max Effort: Not specified	official board Supporting evidence outside the reviewed capability core	2026-10-28
DeepSWE · 1.1	Coding	74.2% Reported settings & source 113 long-horizon tasks; five-trial mean. Section 8.3 does not state reasoning effort. Provider-published lab self-report from the Claude Opus 5.5 system card (SHA256 7311c9c6bbb16d012f1c12c7418b05949fcf7ae3e30d2c40f22050074b2a7378). Not an independent board. Effort is not stated in this section, so it is not copied from the Table 8.1.A max-effort default. Comparison limit: Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison. Claude Opus 5.5 System Card · section 8.3 · reviewed 2026-09-22	Published configuration Effort: Not specified	lab self-report Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison.	Reviewed 2026-09-22
DeepSWE · 1.1	Coding	74.2% Reported settings & source Comparator maximum available thinking/reasoning when reported, otherwise best available result; exact setting not identified by this grid. Argon self-computed with mini-swe agent. Astra from the official public leaderboard; Fable and Opus from their system cards. Highest scoring thinking level per Datacurve. Provider-published launch claim, not independent measurement by Google for every row. Argon self-computed with mini-swe agent. Astra from the official public leaderboard; Fable and Opus from their system cards. Highest scoring thinking level per Datacurve. 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 5: DeepSWE 1.1 / Claude Opus 5.5; 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
FrontierCode · 1.1 Extended	Supporting evidence	65.3% Reported settings & source Cognition agentic coding in Claude Code; composite functional and code-quality score; medium effort; mean@5. Highest Extended score. Cognition ran the evaluation. Provider-published lab self-report from the Claude Opus 5.5 system card (SHA256 7311c9c6bbb16d012f1c12c7418b05949fcf7ae3e30d2c40f22050074b2a7378). Not an independent board. Comparison limit: Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison. Claude Opus 5.5 System Card · section 8.4 · reviewed 2026-09-22	Published configuration Effort: Medium	lab self-report Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison.	Reviewed 2026-09-22
FrontierCode · 1.1 Extended	Supporting evidence	63.6% Reported settings & source Cognition agentic coding in Claude Code; composite functional and code-quality score; max effort; mean@5. Cognition ran the evaluation. Provider-published lab self-report from the Claude Opus 5.5 system card (SHA256 7311c9c6bbb16d012f1c12c7418b05949fcf7ae3e30d2c40f22050074b2a7378). Not an independent board. Comparison limit: Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison. Claude Opus 5.5 System Card · section 8.4 · reviewed 2026-09-22	Published configuration Effort: Max	lab self-report Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison.	Reviewed 2026-09-22
FrontierCode · 1.1 Main	Supporting evidence	54.4% Reported settings & source Cognition agentic coding in Claude Code; composite functional and code-quality score; max effort; mean@5. Cognition ran the evaluation. Provider-published lab self-report from the Claude Opus 5.5 system card (SHA256 7311c9c6bbb16d012f1c12c7418b05949fcf7ae3e30d2c40f22050074b2a7378). Not an independent board. Launch grid shows 54.4% for FrontierCode v1.1 (Main). Section 8.4 says performance falls above medium effort and this max-effort score is 54.4%. Comparison limit: Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison. Claude Opus 5.5 System Card · Table 8.1.A; section 8.4 · reviewed 2026-09-22	Published configuration Effort: Max	lab self-report Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison.	Reviewed 2026-09-22
FrontierCode · 1.1 Main	Supporting evidence	54.6% Reported settings & source Cognition agentic coding in Claude Code; composite functional and code-quality score; medium effort; mean@5. Highest Main score. Cognition ran the evaluation. Provider-published lab self-report from the Claude Opus 5.5 system card (SHA256 7311c9c6bbb16d012f1c12c7418b05949fcf7ae3e30d2c40f22050074b2a7378). Not an independent board. Launch-page prose also states 54.6% at default (medium) effort. Comparison limit: Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison. Claude Opus 5.5 System Card · section 8.4 · reviewed 2026-09-22	Published configuration Effort: Medium	lab self-report Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison.	Reviewed 2026-09-22
FrontierCode 1.1	Supporting evidence	52.17309999999999%	Artificial Analysis FrontierCode 1.1 Stirrup Effort: Not specified	official board Supporting evidence outside the reviewed capability core	2026-10-30
FrontierCS 2	Supporting evidence	58.3%	Claude Code Effort: Not specified	official board Supporting evidence outside the reviewed capability core	2026-10-29
FrontierSWE · 2	Supporting evidence	62.3% Reported settings & source Proximal agent harness; max reasoning effort; 34 tasks; five trials per task; mean across trials. Provider-published lab self-report from the Claude Opus 5.5 system card (SHA256 7311c9c6bbb16d012f1c12c7418b05949fcf7ae3e30d2c40f22050074b2a7378). Not an independent board. Comparison limit: Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board 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. Claude Opus 5.5 System Card · section 8.7 · reviewed 2026-09-22	Published configuration Effort: Max	lab self-report Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board 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 2026-09-22
FrontierSWE · 2	Supporting evidence	62.3% 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 / Claude Opus 5.5; 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	62.3%	Proximal FrontierSWE v2 · Proximus · unspecified Effort: Not specified	official board Supporting evidence outside the reviewed capability core	2026-10-31
gdp.pdf · not specified	Agentic	25.6% 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. Reported combined setup: Opus 5.5 w/ fallbacks. Fallback behavior must not be attributed to the base model alone. Provider-published lab self-report. Published chart cost per task $0.76411812; 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. Comparison limit: Combined fallback setup is not the named base model at one effort setting. Introducing GPT-6.1 Sol · Chart: GDP.pdf / Opus 5.5 w/ fallbacks / low · reviewed 2026-09-29	Published configuration Effort: Low	lab self-report Combined fallback setup is not the named base model at one effort setting.	Reviewed 2026-09-29
gdp.pdf · not specified	Agentic	25.6% 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. Reported combined setup: Opus 5.5 w/ fallbacks. Fallback behavior must not be attributed to the base model alone. Provider-published lab self-report. Published chart cost per task $0.7953176399999999; 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. Comparison limit: Combined fallback setup is not the named base model at one effort setting. Introducing GPT-6.1 Sol · Chart: GDP.pdf / Opus 5.5 w/ fallbacks / medium · reviewed 2026-09-29	Published configuration Effort: Medium	lab self-report Combined fallback setup is not the named base model at one effort setting.	Reviewed 2026-09-29
gdp.pdf · not specified	Agentic	28.800000000000004% 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. Reported combined setup: Opus 5.5 w/ fallbacks. Fallback behavior must not be attributed to the base model alone. Provider-published lab self-report. Published chart cost per task $0.82542564; 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. Comparison limit: Combined fallback setup is not the named base model at one effort setting. Introducing GPT-6.1 Sol · Chart: GDP.pdf / Opus 5.5 w/ fallbacks / high · reviewed 2026-09-29	Published configuration Effort: High	lab self-report Combined fallback setup is not the named base model at one effort setting.	Reviewed 2026-09-29
gdp.pdf · not specified	Agentic	26.6% 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. Reported combined setup: Opus 5.5 w/ fallbacks. Fallback behavior must not be attributed to the base model alone. Provider-published lab self-report. Published chart cost per task $0.9642728800000001; 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. Comparison limit: Combined fallback setup is not the named base model at one effort setting. Introducing GPT-6.1 Sol · Chart: GDP.pdf / Opus 5.5 w/ fallbacks / xhigh · reviewed 2026-09-29	Published configuration Effort: XHigh	lab self-report Combined fallback setup is not the named base model at one effort setting.	Reviewed 2026-09-29
gdp.pdf · not specified	Agentic	26.2% 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. Reported combined setup: Opus 5.5 w/ fallbacks. Fallback behavior must not be attributed to the base model alone. Provider-published lab self-report. Published chart cost per task $1.55233548; 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. Comparison limit: Combined fallback setup is not the named base model at one effort setting. Introducing GPT-6.1 Sol · Chart: GDP.pdf / Opus 5.5 w/ fallbacks / max · reviewed 2026-09-29	Published configuration Effort: Max	lab self-report Combined fallback setup is not the named base model at one effort setting.	Reviewed 2026-09-29
GDPval-AA · 2.1	Agentic	1846 Reported settings & source Artificial Analysis GDPval-AA v2.1; 220 GDPval gold tasks; blind pairwise Elo anchored to DeepSeek V4.1 Flash (max) at 1600; max effort. Run by Artificial Analysis. Provider-published lab self-report from the Claude Opus 5.5 system card (SHA256 7311c9c6bbb16d012f1c12c7418b05949fcf7ae3e30d2c40f22050074b2a7378). Not an independent board. Launch grid shows 1846. Comparison limit: Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison. Claude Opus 5.5 System Card · Table 8.1.A; section 8.14.3 · reviewed 2026-09-22	Published configuration Effort: Max	lab self-report Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison.	Reviewed 2026-09-22
GDPval-AA · 2.1	Agentic	1820 Reported settings & source Artificial Analysis GDPval-AA v2.1; 220 GDPval gold tasks; blind pairwise Elo anchored to DeepSeek V4.1 Flash (max) at 1600; xhigh effort. Run by Artificial Analysis. Provider-published lab self-report from the Claude Opus 5.5 system card (SHA256 7311c9c6bbb16d012f1c12c7418b05949fcf7ae3e30d2c40f22050074b2a7378). Not an independent board. Comparison limit: Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison. Claude Opus 5.5 System Card · section 8.14.3 · reviewed 2026-09-22	Published configuration Effort: XHigh	lab self-report Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison.	Reviewed 2026-09-22
GMMLU	Knowledge	94.3% Reported settings & source Average accuracy across 42 languages; adaptive thinking at max effort; single trial; no tools or custom system prompt. Provider-published lab self-report from the Claude Opus 5.5 system card (SHA256 7311c9c6bbb16d012f1c12c7418b05949fcf7ae3e30d2c40f22050074b2a7378). Not an independent board. Comparison limit: Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison. Claude Opus 5.5 System Card · section 8.16.1; Figure 8.16.1.A · reviewed 2026-09-22	Published configuration Effort: Max	lab self-report Provider-published lab self-report. Shown as a labeled claim and not admitted as a matched board comparison.	Reviewed 2026-09-22
GraphWalks · 256K to 1M BFS F1	Supporting evidence	66.8% 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 / Claude Opus 5.5; 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