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

spatial reasoning benchmark

RefSpatialBench Leaderboard

RefSpatialBench evaluates spatial reference understanding and grounding.

Updated Aug 17, 2026

Models6
Model coverage6
MetricScore
EvidenceB

On this page

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

RefSpatialBench Ranking

Higher score ranks better on this benchmark.

6 rows
Columns

Show columns

Sort by
Rank
Model
Score
Percentile
Participants
Evidence
Evaluated
Rank01ModelACQwen3.6-27BAlibaba Cloud / Qwen TeamScore0.7 pointsPercentile100.0%Participants6EvidenceCEvaluatedAug 17, 2026
Rank02ModelACQwen3 VL 235B A22B ThinkingAlibaba Cloud / Qwen TeamScore0.699 pointsPercentile80.0%Participants6EvidenceCEvaluatedAug 17, 2026
Rank03ModelACQwen3.5-122B-A10BAlibaba Cloud / Qwen TeamScore0.693 pointsPercentile60.0%Participants6EvidenceCEvaluatedAug 17, 2026
Rank04ModelACQwen3.5-27BAlibaba Cloud / Qwen TeamScore0.677 pointsPercentile40.0%Participants6EvidenceCEvaluatedAug 17, 2026
Rank05ModelACQwen3.6-35B-A3BAlibaba Cloud / Qwen TeamScore0.643 pointsPercentile20.0%Participants6EvidenceCEvaluatedAug 17, 2026
Rank06ModelACQwen3.5-35B-A3BAlibaba Cloud / Qwen TeamScore0.635 pointsPercentile0.0%Participants6EvidenceCEvaluatedAug 17, 2026

RefSpatialBench Highlights

The leading models and scores on this benchmark.

Rank #1Qwen3.6-27B0.7 pointsRank #2Qwen3 VL 235B A22B Thinking0.699 pointsRank #3Qwen3.5-122B-A10B0.693 pointsRank #4Qwen3.5-27B0.677 points

RefSpatialBench Score Distribution

A closer view of the leading scores on this benchmark.

RefSpatialBench

The Top AI Models for RefSpatialBench

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

Ranking basisThis refspatialbench 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
    AC
    Qwen3.6-27BAlibaba Cloud / Qwen Team
    Score
    0.7 points
    Price
    $0.60 input / $3.6 output per 1M tokens
    Speed
    Up to 6.1 tok/s via Novita

    Strengths

    • Ranks #1 of 6 compared models
    • 100th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures RefSpatialBench, not total model capability
  2. 02
    AC
    Qwen3 VL 235B A22B ThinkingAlibaba Cloud / Qwen Team
    Score
    0.699 points

    Strengths

    • Ranks #2 of 6 compared models
    • 80th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures RefSpatialBench, not total model capability
  3. 03
    AC
    Qwen3.5-122B-A10BAlibaba Cloud / Qwen Team
    Score
    0.693 points
    Price
    $0.40 input / $3.2 output per 1M tokens

    Strengths

    • Ranks #3 of 6 compared models
    • 60th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures RefSpatialBench, not total model capability
  4. 04
    AC
    Qwen3.5-27BAlibaba Cloud / Qwen Team
    Score
    0.677 points
    Price
    $0.30 input / $2.4 output per 1M tokens
    Speed
    Up to 6.7 tok/s via Novita

    Strengths

    • Ranks #4 of 6 compared models
    • 40th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures RefSpatialBench, not total model capability
  5. 05
    AC
    Qwen3.6-35B-A3BAlibaba Cloud / Qwen Team
    Score
    0.643 points
    Price
    $0.25 input / $1.5 output per 1M tokens

    Strengths

    • Ranks #5 of 6 compared models
    • 20th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures RefSpatialBench, not total model capability

Selection summary

Best AI Models for RefSpatialBench

Qwen3.6-27B currently leads RefSpatialBench with 0.7 points. 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 #1Qwen3.6-27B0.7 points · $0.60 input / $3.6 output per 1M tokensBenchmark rank #2Qwen3 VL 235B A22B Thinking0.699 pointsBenchmark rank #3Qwen3.5-122B-A10B0.693 points · $0.40 input / $3.2 output per 1M tokens

What is RefSpatialBench?

What RefSpatialBench measures and how its scores work.

RefSpatialBench evaluates spatial reference understanding and grounding.

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

Family
RefSpatialBench
Modality
image
Primary category
spatial reasoning
Score direction
higher
LLMBoard eligible
Yes
Evaluation key
refspatialbench|llm-stats-current

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

FAQ

Common questions about RefSpatialBench.

Which model scores highest on RefSpatialBench?

Qwen3.6-27B is currently ranked first with 0.7 points.

What does RefSpatialBench measure?

RefSpatialBench evaluates spatial reference understanding and grounding.

Is a higher score better?

Yes. Higher values rank better for this benchmark.

How many models are compared?

6 model results are currently shown.

Does this benchmark affect the overall score?

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