InclusionAI · Aug 4, 2026
Source confirmed
LLM leaderboard, August 2026
LLM Leaderboard & AI Model Benchmarks — August 2026
Compare frontier AI models by quality, cost, and context. 104 Supported and 111 Estimated models among 378 tracked LLMs — 381 benchmarks, real pricing, and runtime data in one place.
- Data refreshed
- August 4, 2026
- Supported
- 104 models
- Ranking method
- BenchAlign v5.2
Radar · latest confirmed releases
3 source-linked updatesLiquidAI · Aug 4, 2026
Source confirmed
Alibaba · Aug 3, 2026
Source confirmed
New model, price change, API update, or outage: Radar sends what matters to your inbox with the source. Try it free for seven days; plans start at $4.99/month.
AI moves fast. Don’t be the last to know.
New models, price changes, benchmark moves, and the releases worth waiting for—one concise email each week.
Read a sample issueJoin 2,000+ readers.
One email each week. Unsubscribe anytime.
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.MiniMax M3MiniMax68.8overall scoreSupported · 2 source families
- Best near-frontier value91% of the leading score at a lower output price.Grok 4.5xAI$6output / 1M tokensSupported · 3 source families
- Fastest measuredFastest externally measured model that also clears the ranking and evidence thresholds.Nemotron 3 Nano Omni 30B A3BNVIDIA323tokens / secArtificial Analysis · updated 2026-08-04
- Largest useful contextLargest context among evidence-qualified current models retaining at least half of the leading score.Grok 4.20xAI2Mcontext windowEstimated · 1 source family
The LLM leaderboard — August 2026 rankings
Benchmarks, pricing, runtime signals, and context window 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.
378 models
Every row can rank. Supported positions have diverse direct evidence; Estimated positions remain visible with wider uncertainty.Supporteddiverse direct evidenceEstimatedlimited direct coverage
1 Claude Mythos 5 AnthropicSupported | Anthropic | Closed | Current | Reasoning | 1M+ | $10.00 / $50.00 | Not measured | Not measured | 83.04 | 76 | 80 | — | 86 | 94 | — | — | — | — |
2 Claude Fable 5 AnthropicSupported | Anthropic | Closed | Current | Reasoning | 1M+ | $10.00 / $50.00 | 73 | 83.30s | 82.79 | 75 | 80 | — | 63 | 71 | — | — | — | 1508.58 |
3 Claude Opus 5 AnthropicSupported | Anthropic | Closed | Current | Reasoning | — | $5.00 / $25.00 | 56 | 63.43s | 82.59 | 81 | 77 | 91 | 89 | 94 | — | — | — | 1491.82 |
4 GPT-5.6 Sol OpenAISupported | OpenAI | Closed | Current | Reasoning | 1.05M | $5.00 / $30.00 | 71 | 131.39s | 81.48 | 68 | 78 | 93 | 85 | 83 | — | — | 97 | 1482.77 |
5 Kimi K3 Moonshot AISupported | Moonshot AI | Closed | Current | Reasoning | 1.05M | $3.00 / $15.00 | 36 | 58.35s | 79.89 | 74 | 77 | — | 88 | 85 | — | — | — | 1485.33 |
6 Claude Opus 4.8 AnthropicSupported | Anthropic | Closed | Superseded | Reasoning | 1M | $5.00 / $25.00 | 58 | 22.68s | 77.34 | 61 | 72 | 76 | 88 | 88 | — | — | 67 | 1483.64 |
7 Muse Spark 1.1 MetaSupported | Meta | Closed | Current | Reasoning | 1M | Not listed | 216 | 11.92s | 76.15 | 65 | 64 | — | 76 | 94 | — | — | — | 1489.67 |
8 Grok 4.5 xAISupported | xAI | Closed | Current | Reasoning | 500K | $2.00 / $6.00 | 61 | 10.38s | 75.38 | 61 | 57 | 60 | 68 | 69 | — | — | — | 1468.83 |
9 Gemini 3.6 Flash GoogleSupported | Closed | Current | Reasoning | 1M | $1.50 / $7.50 | 213 | 16.09s | 75.3 | 47 | 64 | — | 74 | 70 | — | — | — | 1482.68 | |
10 GPT-5.4 OpenAISupported | OpenAI | Closed | Superseded | Reasoning | 1.05M | $2.50 / $15.00 | 143 | 130.50s | 73.2 | 57 | 58 | 78 | 67 | 77 | — | — | 66 | 1465.37 |
11 GPT-5.5 OpenAIEstimated | OpenAI | Closed | Superseded | Reasoning | 1M | $5.00 / $30.00 | 87 | 65.57s | 72.28 | 64 | 71 | 87 | 68 | 79 | — | — | 71 | 1482.44 |
12 GPT-5.6 Terra OpenAIEstimated | OpenAI | Closed | Current | Reasoning | 1.05M | $2.00 / $12.00 | Not measured | Not measured | 72.26 | 63 | 65 | 86 | 75 | 82 | — | — | 97 | — |
13 Qwen3.7 Max AlibabaSupported | Alibaba | Closed | Superseded | Reasoning | 1M | Not listed | 205 | 14.07s | 71.8 | 40 | 54 | 87 | — | 72 | 100 | 93 | 82 | 1474.66 |
14 Claude Opus 4.7 (Adaptive) AnthropicEstimated | Anthropic | Closed | Superseded | Reasoning | 1M | $5.00 / $25.00 | Not measured | Not measured | 71.46 | 63 | 64 | 79 | 48 | 83 | — | — | — | 1501.5 |
15 Claude Opus 4.7 AnthropicSupported | Anthropic | Closed | Current | Standard | 1M | $5.00 / $25.00 | 52 | 22.82s | 71.21 | 56 | 67 | — | — | 64 | — | — | 62 | 1492.41 |
16 Muse Spark MetaSupported | Meta | Closed | Superseded | Reasoning | 262K | Not listed | Not measured | Not measured | 70.49 | 58 | 62 | 52 | 76 | 75 | — | — | 56 | 1487.88 |
17 MiMo-V2.5-Pro XiaomiSupported | Xiaomi | Closed | Current | Reasoning | 1M | Not listed | Not measured | Not measured | 69.21 | 42 | 55 | — | — | 71 | — | — | — | 1465.71 |
18 | MiniMax | Open | Current | Standard | 1M | $0.30 / $1.20 | 92 | 23.04s | 68.8 | 40 | 48 | — | 48 | 66 | — | — | — | 1445.27 |
| Tencent | Open | Current | Reasoning | 256K | $0.00 / $0.00 | 70 | 31.36s | 67.91 | 61 | 63 | — | — | — | — | — | — | 1456.27 | |
20 Claude Opus 4.6 AnthropicSupported | Anthropic | Closed | Superseded | Standard | 1M | $5.00 / $25.00 | 37 | 2.51s | 67.74 | 56 | 61 | — | 62 | 83 | — | — | 60 | 1497.19 |
21 Gemini 3 Pro GoogleSupported | Closed | Established | Standard | 2M | $2.00 / $12.00 | 109 | 32.65s | 66.96 | 58 | 61 | 42 | 73 | 69 | — | — | 56 | 1485.73 | |
| Z.AI | Open | Superseded | Reasoning | 203K | $1.40 / $4.40 | 73 | 53.82s | 66.9 | 49 | 55 | — | — | 78 | — | — | 65 | 1468.79 | |
23 GPT-5.6 Luna OpenAIEstimated | OpenAI | Closed | Current | Reasoning | 1.05M | $0.20 / $1.20 | Not measured | Not measured | 66.85 | 56 | 72 | 66 | 66 | 81 | — | — | 97 | — |
24 MiMo-V2-Pro XiaomiSupported | Xiaomi | Closed | Superseded | Reasoning | 1M | Not listed | Not measured | Not measured | 66.76 | — | 61 | — | — | — | — | — | — | 1448.24 |
| Thinking Machines Lab | Open | Current | Standard | 1M | $1.87 / $4.68 | 85 | 25.24s | 66.5 | 49 | 43 | — | 51 | 69 | — | 91 | 79 | 1440.8 |
Showing 25 of 378
The BenchLM LLM leaderboard 2026 ranks 215+ models and tracks 378+ large language models side by side across 381 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.
Dataset journal
What changed in August 2026
- Claude Mythos 5 (Anthropic) holds the #1 spot on the BenchAlign leaderboard at 83.04.
- 4 model releases tracked in August 2026.
- 104 Supported and 111 Estimated models appear in the public leaderboard.
Data verified
Build a head-to-head comparison
Choose any two ranked models. The comparison opens with the decision, exact units, and evidence coverage.
versusCompare these models
The evidence map
Explore benchmarks by category
Move from the overall ranking into the capability or workload that actually decides your model choice.
Agentic82Tools, browsers, and long-running workCoding54Repository tasks and code generationMultimodal65Documents, images, and grounded workReasoning26Long-context and abstract reasoningKnowledge46Professional and scientific knowledgeInstruction following4Constraint and format reliabilityMultilingual13Cross-language accuracy and coverageMath32Contest and frontier mathematics
More ranking paths
Choose by workload
Long ContextLongBench v2, MRCR, AI-Needle, GraphwalksTool UseBFCL, MCP Atlas, Toolathlon, TAU BenchWeb ResearchBrowseComp, WebArena, WebVoyagerComputer UseOSWorld, ScreenSpot Pro, Vision2WebDocument AIOfficeQA Pro, OmniDocBench, CC-OCRImage UnderstandingMMMU-Pro, AI2D, CountBench, RefCOCOFrontend / App DevReact Native Evals, Design2Code, Vision2WebFactualitySimpleQA, HLE w/o tools, Facts-VLM
Six-month release record
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.
- Months
- 6
- Releases
- 155
- Lead changes
- 5
Best ranked model released in Aug 2026
Qwen3.8 MaxAlibabaScore60.9
RankProviderTop-three avg.Models
01Anthropic82.820
02OpenAI75.738
03Google69.117
Popular comparisons
Tools and resources
AI Prompt WriterTurn the work you need into a testable promptPrompt OptimizerGet closer to usable output with a clearer promptLLM Pricing ComparisonCompare API costs across modelsAlternative FinderFind the best ChatGPT or Claude replacementWhich LLM Should I Use?Get a personalized recommendationCost CalculatorEstimate your monthly API spendMethodology & ExplainersHow BenchLM scores and curates benchmarks
Scoring methodology
Scoring methodology8 weighted categories, 27 ranking benchmarks, and 295 display-only records.Inspect method
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.
Data comes from OpenBench, official model papers, and public leaderboards. External consensus signals are bounded calibration inputs and are not exposed in exported data.
Read the complete methodology- Category weights
- 100%
- Scored categories
- 8
- Ranking benchmarks
- 27
- Display-only
- 295
Category and weightBenchmarks that affect the scoreCoverage
- Agentic22%
Terminal-Bench 2.0 · BrowseComp · OSWorld-Verified
3 weighted · 79 context-only
- Coding20%
SWE-bench Verified · SWE-Rebench · LiveCodeBench · SWE-bench Pro · SciCode
5 weighted · 49 context-only
- Reasoning17%
LongBench v2 · MRCRv2 · ARC-AGI-2
3 weighted · 23 context-only
- Multimodal12%
MMMU-Pro · OfficeQA Pro · CharXiv
3 weighted · 62 context-only
- Knowledge12%
GPQA · SuperGPQA · MMLU-Pro · HLE · SimpleQA
5 weighted · 41 context-only
- Multilingual7%
MMLU-ProX
1 weighted · 12 context-only
- Instruction Following5%
IFEval · IFBench
2 weighted · 2 context-only
- Math5%
AIME26 · HMMT Feb 2026 · FrontierMath v2 (Tiers 1-3) · FrontierMath v2 (Tier 4) · USAMO 2026
5 weighted · 27 context-only