LLM Leaderboard & AI Model Benchmarks — November 2026

As of November 10, 2026, the BenchLM AI leaderboard ranks 214 models. Claude Fable 5.2 is first by point estimate. Conditional ranges do not establish rank confidence. Open the evidence before choosing.

Search models and benchmarks

Current BenchAlign leaders

Snapshot November 10, 2026

Method

Claude Fable 5.2Estimated
Anthropic01
Score92.05Conditional range 77.7–100.0
Agentic
90
Coding
87
Reasoning
90
Multimodal
90
Knowledge
98
Multilingual
Score pending
Instruction following
Not measured
Math
94not ranked
Estimatedwider uncertainty
Details for Claude Fable 5.2
Gemini 4 CarbonEstimated
Google02
Score88.08Conditional range 78.4–97.8
Agentic
81not ranked
Coding
81
Reasoning
78not ranked
Multimodal
Score pending
Knowledge
Not measured
Multilingual
Not measured
Instruction following
Score pending
Math
Not measured
Estimatedwider uncertainty
Details for Gemini 4 Carbon
Claude Opus 5.5Supported
Anthropic03
Score85.7490% interval 80.9–90.6
Agentic
89
Coding
82
Reasoning
80
Multimodal
84
Knowledge
88
Multilingual
Score pending
Instruction following
Score pending
Math
93not ranked
Supportedenough independent evidence
Details for Claude Opus 5.5

Price context: GPT-6.1 Sol retains 92% of the top score with an output price 75% lower.

Data refreshed: November 10, 2026
Supported: 78 models
Ranking method: BenchAlign v5.8

The LLM leaderboard — November 2026 rankings

Score, model size, context, price, and benchmark detail in one table. Filter state syncs to the URL so every view is shareable. Supported and Estimated labels show how much independent evidence backs each position.

Methodology Providers AI Race

CSV API/MCP

Supported means we have enough independent evidence to stand behind the row. Estimated means the rank is visible and the uncertainty is wider.

Supportedenough independent evidenceEstimatedwider uncertainty

Parameters show total billions; active counts appear when a source reports them. Size filters use total parameters: Tiny ≤4B, Small >4–<40B, Medium 40–150B, and Large >150B.

AG Agentic CO Coding RE Reasoning MM Multimodal & Grounded KN Knowledge ML Multilingual IF Instruction Following MA Math

LLM leaderboard table

#1 Claude Fable 5.2 AnthropicEstimated	92.05	Not disclosed	1M	$8.00 / $40.00	—	Closed	90	87	90	90	98	—	—	—	Not measured	Not measured	Proprietary	Current	Reasoning
#2 Gemini 4 Carbon GoogleEstimated	88.08	Not disclosed	1M	$2.00 / $10.00	1525.95	Closed	—	81	—	—	—	—	—	—	100	97.48s	Proprietary	Current	Reasoning
#3 Claude Opus 5.5 AnthropicSupported	85.74	Not disclosed	1M	$4.00 / $20.00	1506.07	Closed	89	82	80	84	88	—	—	—	98	663.66s	Proprietary	Current	Reasoning
#4 GPT-6 Astra OpenAISupported	85.18	Not disclosed	1.05M	$10.00 / $50.00	1474.09	Closed	74	74	81	79	84	—	—	—	73	250.99s	Proprietary	Current	Reasoning
#5 GPT-6.1 Sol OpenAISupported	85	Not disclosed	1.05M	$2.00 / $10.00	1482.07	Closed	76	69	80	78	84	—	—	—	Not measured	Not measured	Proprietary	Current	Reasoning
#6 Gemini 4 Argon GoogleSupported	81.54	Not disclosed	1M	$2.00 / $10.00	1522.38	Closed	78	72	78	—	79	—	—	—	83	44.96s	Proprietary	Superseded	Reasoning
#7 Claude Sonnet 5.5 AnthropicSupported	81.23	Not disclosed	1M	$2.00 / $10.00	1472.49	Closed	68	81	76	78	80	—	—	—	138	448.12s	Proprietary	Current	Reasoning
#8 Claude Fable 5.1 AnthropicSupported	80.9	Not disclosed	1M	$10.00 / $50.00	1502.7	Closed	80	76	77	81	85	—	—	—	66	128.91s	Proprietary	Superseded	Reasoning
#9 Claude Fable 5 AnthropicSupported	80.33	Not disclosed	1M+	$10.00 / $50.00	1503.91	Closed	74	71	72	75	80	—	77	—	68	100.13s	Proprietary	Superseded	Reasoning
#10 GPT-6.1 Terra OpenAIEstimated	80.09	Not disclosed	1.05M	$1.00 / $5.00	—	Closed	76	—	75	77	76	—	—	—	Not measured	Not measured	Proprietary	Current	Reasoning
#11 Claude Opus 5 AnthropicSupported	79.08	Not disclosed	—	$5.00 / $25.00	1488.79	Closed	77	70	75	78	80	—	—	—	62	59.73s	Proprietary	Superseded	Reasoning
#12 GPT-6 Sol OpenAISupported	77.74	Not disclosed	1.05M	$2.00 / $10.00	1456.83	Closed	58	65	72	75	79	—	—	—	Not measured	Not measured	Proprietary	Superseded	Reasoning
#13 Kimi K3.1 Pro Moonshot AIEstimated Self-host	74.59	2800B104B active	1.05M	$2.00 / $10.00	1483.02	Open	73	—	71	74	74	—	—	—	43	49.60s	Open Weight	Current	Reasoning
#14 GPT-5.6 Sol OpenAISupported	74.34	Not disclosed	1.05M	$4.00 / $20.00	1484.95	Closed	68	67	74	75	75	—	89	—	94	100.66s	Proprietary	Superseded	Reasoning
#15 GPT-5.5 Pro OpenAIEstimated	72.46	Not disclosed	1M	$30.00 / $180.00	—	Closed	—	—	75	—	—	—	—	—	Not measured	Not measured	Proprietary	Established	Reasoning
#16 GPT-5.6 Terra OpenAISupported	69.41	Not disclosed	1.05M	$2.00 / $12.00	1467.87	Closed	60	62	69	69	68	—	87	—	Not measured	Not measured	Proprietary	Superseded	Reasoning
#17 Kimi K3 Moonshot AISupported Self-host	69.28	Not disclosed	1.05M	$3.00 / $15.00	1487.12	Open	67	59	65	67	67	—	—	—	38	56.25s	Open Weight	Superseded	Reasoning
#18 Claude Opus 4.8 AnthropicSupported	69.18	Not disclosed	1M	$5.00 / $25.00	1474.26	Closed	60	60	65	67	69	—	75	79	65	35.96s	Proprietary	Superseded	Reasoning
#19 Muse Wonder Full MetaEstimated	69.14	Not disclosed	1M	Not listed	—	Closed	—	71	76	—	74	—	—	—	Not measured	Not measured	Proprietary	Current	Reasoning
#20 GPT-5.5 OpenAISupported	68.35	Not disclosed	1M	$5.00 / $30.00	1481.31	Closed	60	62	67	68	70	—	93	—	98	37.05s	Proprietary	Superseded	Reasoning
#21 Claude Opus 4.7 (Adaptive) AnthropicEstimated	68	Not disclosed	1M	$5.00 / $25.00	1502.21	Closed	57	56	—	62	65	—	—	—	Not measured	Not measured	Proprietary	Superseded	Reasoning
#22 GPT-5.4 Pro OpenAIEstimated	67.87	Not disclosed	1.05M	$30.00 / $180.00	1475.23	Closed	—	—	73	—	—	—	—	—	74	151.79s	Proprietary	Established	Reasoning
#23 Gemini 3.8 Flash GoogleSupported	66.69	Not disclosed	1M	$0.75 / $3.75	1497.49	Closed	65	60	65	70	69	—	—	—	140	27.53s	Proprietary	Current	Reasoning
#24 Grok 4.6 xAISupported	66.55	Not disclosed	500K	$2.00 / $6.00	1456.49	Closed	70	59	65	—	68	—	—	—	61	35.43s	Proprietary	Superseded	Reasoning
#25 Gemini 3.7 Flash GoogleSupported	66.28	Not disclosed	1M	$0.75 / $3.75	1489.39	Closed	61	58	63	70	70	—	—	—	285	8.52s	Proprietary	Superseded	Reasoning

Showing 25 of 1020 models · 214 ranked

Source-linked · the table lists all 1020 tracked models; 214 carry a BenchAlign rank

How the score is built

Ask your AI, from current BenchLM data

Which models lead the BenchLM ranking right now, and what do they cost?

Connect BenchLM to your AI

What the table covers

The BenchLM LLM leaderboard 2026 ranks 214+ models and tracks 1020+ large language models side by side across 811 benchmarks — from SWE-bench and LiveCodeBench for coding to GPQA Diamond and MMLU-Pro for knowledge and reasoning. Whether you need the best AI models 2026 has to offer for agentic workflows, math, multilingual tasks, or instruction following, our AI benchmark comparison tables make it easy to see how GPT-5, Claude, Gemini, DeepSeek, Llama, and dozens of other frontier and open-source models stack up on both benchmarks and operator tradeoffs like price and context. Supported and Estimated labels separate evidence strength from score, so incomplete reporting does not automatically remove a model from the comparison.

Reading the table

Embed the live leaderboard

What changed in November 2026

Claude Fable 5.2 (Anthropic) holds the #1 spot on the BenchAlign leaderboard at 92.05.
10 model releases tracked in November 2026.
Anthropic climbed 1 spot in the provider race.
78 Supported and 136 Estimated models appear in the public leaderboard.

Dataset journalData verified November 10, 2026

Explore benchmarks by category

Move from the overall ranking into the capability or workload that actually decides your model choice.

The evidence mapBenchmark directory

Build a head-to-head comparison

Choose any two ranked models. The comparison opens with the decision, exact units, and evidence coverage.

versus

Compare these models

Decision-ready picks

Current models that clear the ranking, freshness, and evidence thresholds for each decision.

DecisionModelCurrent measureProvenance

Best open weightHighest-scoring evidence-qualified current open-weight model.Kimi K3.1 ProMoonshot AI74.59overall scoreEstimated · 1 source family
Best near-frontier value92% of the leading score at a lower output price.GPT-6.1 SolOpenAI$10output / 1M tokensSupported · 3 source families
Fastest measuredFastest externally measured model that also clears the ranking and evidence thresholds.Gemini 3.5 Flash-LiteGoogle461tokens / secArtificial Analysis · updated 2026-11-10
Largest useful contextLargest context among evidence-qualified current models retaining at least half of the leading score.GPT-6 AstraOpenAI1.05Mcontext windowSupported · 3 source families

Current models that clear rank, freshness and evidence thresholds

Qualification method Qualification method

Popular comparisons

Ornith-1.5-122B vs Ornith-1.5-397B Ornith-1.5-122B vs Ornith-1.5-35B-A3B ScaleDown Classify vs ScaleDown Extract ScaleDown Classify vs ScaleDown Summarize ScaleDown Classify vs ScaleDown Compress ScaleDown Extract vs ScaleDown Summarize ScaleDown Extract vs ScaleDown Compress ScaleDown Summarize vs ScaleDown Compress Gemini 4 Argon vs GPT-6 Astra

Tools and resources

AI Prompt WriterTurn the work you need into a testable prompt Prompt OptimizerGet closer to usable output with a clearer prompt LLM Pricing ComparisonCompare API costs across models Alternative FinderFind the best ChatGPT or Claude replacement Which LLM Should I Use?Get a personalized recommendation Cost CalculatorEstimate your monthly API spend Methodology & ExplainersHow benchmarks are scored and curated

How we work

A ranking is a shortlist.

Rank orders the evidence; it does not make the decision. Every row here opens to its sources so you can disagree with us.

We do not invent scores to fill a table.

A model without a public, independent row stays unranked, and says so. Vendor-run numbers are shown, labeled, and kept out of the rank.

When we were wrong, the fix is on the page.

Scoring bugs, missing coverage, and re-weightings are dated and linked, because a ranking you cannot audit is a launch post with a table.

The desk

Editorial policy

The AI race, read as a record instead of a spectacle

Track which release led its month, how provider depth changed, and whether the crown moved.

Explore the release timeline

Months: 6
Releases: 292
Lead changes: 5

Best ranked model released in Nov 2026

Claude Fable 5.2Anthropic

Score92.05

RankProviderTop-three avg.Models

01Anthropic86.323

02OpenAI83.437

03Google78.820

Six-month release record

Explore the release timeline

Scoring methodology8 weighted categories, 63 ranking benchmarks, and 400 display-only records.Inspect method

Category weights: 100%
Scored categories: 8
Ranking benchmarks: 63
Display-only: 400

Each model's overall score is a normalized weighted average of category averages. Within each category, benchmark results are normalized to a common scale and combined using weights that favor harder, less-saturated evaluations.

Confidence remains separate from score. It shows how much public benchmark evidence supports a result, while display-only benchmarks stay visible for context without receiving ranking weight.

Benchmark scores come from official provider tables, model cards, release announcements, and benchmark-native public leaderboards. External consensus signals are bounded calibration inputs and are not exposed in exported data.

Read the complete methodology

Category and weightBenchmarks that affect the scoreCoverage

Agentic22%
Terminal-Bench 3.0 · Terminal-Bench 4.0 · AA AutomationBench · AA EnterpriseOps-Gym · AA Harvey LAB v1.0 · AA Tau3 Banking · Terminal-Bench 2.0 · Terminal-Bench 2.1 · BrowseComp · HLE w/ tools · GDPval-AA · OSWorld-Verified · JobBench · MCP Atlas · Toolathlon · Toolathlon-Verified · AutomationBench · Agents' Last Exam · DeepSearchQA · BFCL v4 · τ³-bench results · Terminal-Bench 2.1 (Vals)
22 weighted · 123 context-only
Coding20%
SWE-Rebench · SWE-bench Pro · VulcanBench v3 · FrontierCode 1.1 Main · SWE Multilingual · FrontierSWE v2 · SciCode · AA-SciCode · LiveCodeBench (Vals) · DeepSWE
10 weighted · 59 context-only
Reasoning17%
LongBench v2 · MRCRv2 · ARC-AGI-2 · ARC-AGI-3 · AA-LCR · CritPt
6 weighted · 33 context-only
Multimodal12%
ParseBench · MMMU-Pro · AA-MMMU-Pro · OfficeQA Pro · ScreenSpot Pro · CharXiv · CharXiv w/o tools
7 weighted · 71 context-only
Knowledge12%
GPQA · SuperGPQA · MMLU-Pro · HLE · AA-GPQA Diamond · AA-HLE · AA-Omniscience Accuracy · HLE w/o tools · GPQA Diamond (Vals) · MMLU-Pro (Vals)
10 weighted · 57 context-only
Multilingual7%
MMLU-ProX
1 weighted · 13 context-only
Instruction Following5%
IFBench · AA-IFBench
2 weighted · 15 context-only
Math5%
AIME26 · HMMT Feb 2026 · FrontierMath v2 (Tiers 1-3) · FrontierMath v2 (Tier 4) · USAMO 2026
5 weighted · 29 context-only

Scoring methodology · BenchAlign v5.8

Inspect method