llmboard.aiAI model intelligence
Home

Model Rankings

OverallOpen ModelsAgentCodingReasoningMathKnowledgeInstruction FollowingTextVision
Image GenerationImage Editing
Video GenerationImage to VideoVideo Editing
Text to SpeechSpeech to Text
Embeddings

Efficiency

Chat Token PricingImage PricingVideo PricingAudio Pricing
Chat Speed & LatencyProvider Reliability

Benchmarks

GPQAMMLU-ProAIME 2025SWE-Bench VerifiedMMLUHumanity's Last ExamLiveCodeBenchMATHHumanEvalMMMU-Pro
All Benchmarks

Tools

Model DirectoryCompare Models

Scoring & Data

Scoring & Data
393 models668 benchmarks

Leaderboard Center

Overall RankingCodingCore BenchmarksPrice & ValueRuntime Performance

Modalities

All ModelsImage GenerationImage EditingVideo GenerationImage-to-VideoVideo EditingText-to-SpeechSpeech-to-TextEmbeddings

Data & Methods

Scoring MethodAll BenchmarksReasoningMath

Vendors

All VendorsOpenAIAnthropicGoogle
llmboard.aiCopyright 2026 llmboard.ai

safety benchmark

XSTest Leaderboard

XSTest is a test suite designed to identify exaggerated safety behaviours in large language models. It comprises 450 prompts: 250 safe prompts across ten prompt types that well-calibrated models should not refuse to comply with, and 200 unsafe prompts as contrasts that models should refuse. The benchmark systematically evaluates whether models refuse to respond to clearly safe prompts due to overly cautious safety mechanisms.

Updated Aug 17, 2026

Models4
Model coverage4
MetricScore
EvidenceB

On this page

  • Ranking
  • Highlights
  • Distribution
  • Top models
  • About
  • FAQ

XSTest Ranking

Higher score ranks better on this benchmark.

4 rows
Columns

Show columns

Sort by
Rank
Model
Score
Percentile
Participants
Evidence
Evaluated
Rank01ModelGOGemini 1.5 ProGoogleScore98.8%Percentile100.0%Participants4EvidenceCEvaluatedAug 17, 2026
Rank02ModelGOGemini 1.5 FlashGoogleScore97.0%Percentile66.7%Participants4EvidenceCEvaluatedAug 17, 2026
Rank03ModelMAShieldstral 1.0 (3B)Mistral AIScore94.6%Percentile33.3%Participants4EvidenceCEvaluatedAug 17, 2026
Rank04ModelGOGemini 1.5 Flash 8BGoogleScore92.6%Percentile0.0%Participants4EvidenceCEvaluatedAug 17, 2026

XSTest Highlights

The leading models and scores on this benchmark.

Rank #1Gemini 1.5 Pro98.8%Rank #2Gemini 1.5 Flash97.0%Rank #3Shieldstral 1.0 (3B)94.6%Rank #4Gemini 1.5 Flash 8B92.6%

XSTest Score Distribution

A closer view of the leading scores on this benchmark.

XSTest

The Top AI Models for XSTest

The first five results on this benchmark, with official price and output speed added where the model identity can be matched.

Ranking basisThis xstest AI model leaderboard uses descending score in the benchmark's original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.

  1. 01
    GO
    Gemini 1.5 ProGoogle
    Score
    98.8%
    Speed
    Up to 85 tok/s via Google

    Strengths

    • Ranks #1 of 4 compared models
    • 100th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures XSTest, not total model capability
  2. 02
    GO
    Gemini 1.5 FlashGoogle
    Score
    97.0%
    Speed
    Up to 150 tok/s via Google

    Strengths

    • Ranks #2 of 4 compared models
    • 67th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures XSTest, not total model capability
  3. 03
    MA
    Shieldstral 1.0 (3B)Mistral AI
    Score
    94.6%

    Strengths

    • Ranks #3 of 4 compared models
    • 33th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures XSTest, not total model capability
  4. 04
    GO
    Gemini 1.5 Flash 8BGoogle
    Score
    92.6%
    Speed
    Up to 150 tok/s via Google

    Strengths

    • Ranks #4 of 4 compared models
    • 0th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures XSTest, not total model capability

Selection summary

Best AI Models for XSTest

Gemini 1.5 Pro currently leads XSTest with 98.8%. It is the top model on this specific benchmark, while the best LLM for the broader task should also be checked against other benchmarks, price and runtime.

Use this leaderboard with the supporting benchmark results and coverage details above. A leaderboard position summarizes the selected ranking signal; it does not replace workload-specific testing.

Benchmark rank #1Gemini 1.5 Pro98.8% · Up to 85 tok/s via GoogleBenchmark rank #2Gemini 1.5 Flash97.0% · Up to 150 tok/s via GoogleBenchmark rank #3Shieldstral 1.0 (3B)94.6%

What is XSTest?

What XSTest measures and how its scores work.

XSTest is a test suite designed to identify exaggerated safety behaviours in large language models. It comprises 450 prompts: 250 safe prompts across ten prompt types that well-calibrated models should not refuse to comply with, and 200 unsafe prompts as contrasts that models should refuse. The benchmark systematically evaluates whether models refuse to respond to clearly safe prompts due to overly cautious safety mechanisms.

Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.

Family
XSTest
Modality
text
Primary category
safety
Score direction
higher
LLMBoard eligible
Yes
Evaluation key
xstest|llm-stats-current

Benchmark scores retain their original unit. Overall score eligibility is shown separately.

FAQ

Common questions about XSTest.

Which model scores highest on XSTest?

Gemini 1.5 Pro is currently ranked first with 98.8%.

What does XSTest measure?

XSTest is a test suite designed to identify exaggerated safety behaviours in large language models. It comprises 450 prompts: 250 safe prompts across ten prompt types that well-calibrated models should not refuse to comply with, and 200 unsafe prompts as contrasts that models should refuse. The benchmark systematically evaluates whether models refuse to respond to clearly safe prompts due to overly cautious safety mechanisms.

Is a higher score better?

Yes. Higher values rank better for this benchmark.

How many models are compared?

4 model results are currently shown.

Does this benchmark affect the overall score?

Yes. This benchmark can contribute to the current LLMBoard capability score.