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 Directory

Scoring & Data

Scoring & Data
1224 models729 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 Benchmarks
llmboard.aiCopyright 2026 llmboard.ai

long context benchmark

MRCR v2 (8-needle) Leaderboard

MRCR v2 (8-needle) is a variant of the Multi-Round Coreference Resolution benchmark with 8 needle items to retrieve from long contexts.

Updated Sep 5, 2026

Models24
Model coverage24
MetricScore
EvidenceC

On this page

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

MRCR v2 (8-needle) Ranking

Higher score ranks better on this benchmark.

24 rows
Columns

Show columns

Sort by
Rank
Model
Score
Percentile
Participants
Evidence
Evaluated
Rank01ModelMEMuse Spark 1.3MetaScore98.50%Percentile100.00%Participants24EvidenceCEvaluatedSep 8, 2026
Rank02ModelGOGemini 3.7 FlashGoogleScore97.00%Percentile95.65%Participants24EvidenceCEvaluatedSep 8, 2026
Rank03ModelACQwen3.8 MaxAlibaba Cloud / Qwen TeamScore92.90%Percentile91.30%Participants24EvidenceCEvaluatedSep 8, 2026
Rank04ModelOPGPT-5.6 SolOpenAIScore91.50%Percentile86.96%Participants24EvidenceCEvaluatedSep 8, 2026
Rank05ModelOPGPT-5.6 TerraOpenAIScore89.60%Percentile82.61%Participants24EvidenceCEvaluatedSep 8, 2026
Rank06ModelANClaude Opus 4.6AnthropicScore76.00%Percentile78.26%Participants24EvidenceCEvaluatedSep 8, 2026
Rank07ModelOPGPT-5.5OpenAIScore74.00%Percentile73.91%Participants24EvidenceCEvaluatedSep 8, 2026
Rank08ModelGOGemma 4 31BGoogleScore66.40%Percentile69.57%Participants24EvidenceCEvaluatedSep 8, 2026
Rank09ModelGOGemini 3.1 Flash-LiteGoogleScore60.10%Percentile65.22%Participants24EvidenceCEvaluatedSep 8, 2026
Rank10ModelGOGemini 3.6 FlashGoogleScore54.00%Percentile60.87%Participants24EvidenceCEvaluatedSep 8, 2026
Rank11ModelGOGemma 4 26B-A4BGoogleScore44.10%Percentile56.52%Participants24EvidenceCEvaluatedSep 8, 2026
Rank12ModelGOGemma 4 12BGoogleScore43.40%Percentile52.17%Participants24EvidenceCEvaluatedSep 8, 2026
Rank13ModelOPGPT-5.6 LunaOpenAIScore41.30%Percentile47.83%Participants24EvidenceCEvaluatedSep 8, 2026
Rank14ModelOPGPT-5.4 miniOpenAIScore33.60%Percentile43.48%Participants24EvidenceCEvaluatedSep 8, 2026
Rank15ModelOPGPT-5.4 nanoOpenAIScore33.10%Percentile39.13%Participants24EvidenceCEvaluatedSep 8, 2026
Rank16ModelGOGemini 3.5 FlashGoogleScore26.60%Percentile34.78%Participants24EvidenceCEvaluatedSep 8, 2026
Rank17ModelGOGemini 3.1 ProGoogleScore26.30%Percentile30.43%Participants24EvidenceCEvaluatedSep 8, 2026
Rank18ModelGOGemini 3 ProGoogleScore26.30%Percentile26.09%Participants24EvidenceCEvaluatedSep 8, 2026
Rank19ModelGOGemma 4 E4BGoogleScore25.40%Percentile21.74%Participants24EvidenceCEvaluatedSep 8, 2026
Rank20ModelGOGemini 3 FlashGoogleScore22.10%Percentile17.39%Participants24EvidenceCEvaluatedSep 8, 2026
Rank21ModelGOGemini 3.5 Flash-LiteGoogleScore21.30%Percentile13.04%Participants24EvidenceCEvaluatedSep 8, 2026
Rank22ModelGOGemma 4 E2BGoogleScore19.10%Percentile8.70%Participants24EvidenceCEvaluatedSep 8, 2026
Rank23ModelGOGemini 2.5 Pro Preview 06-05GoogleScore16.40%Percentile4.35%Participants24EvidenceCEvaluatedSep 8, 2026
Rank24ModelGOGemma 3 27BGoogleScore13.50%Percentile0.00%Participants24EvidenceCEvaluatedSep 8, 2026

MRCR v2 (8-needle) Highlights

The leading models and scores on this benchmark.

MRCR v2 (8-needle) Score Distribution

A closer view of the leading scores on this benchmark.

MRCR v2 (8-needle)

The Top AI Models for MRCR v2 (8-needle)

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

What is MRCR v2 (8-needle)?

What MRCR v2 (8-needle) measures and how its scores work.

MRCR v2 (8-needle) is a long_context benchmark variant of the Multi-Round Coreference Resolution benchmark.

It measures models' ability to simultaneously track and reason about multiple pieces of information across extended conversations.

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

Family
MRCR v2 (8-needle)
Modality
text
Primary category
long context
Score direction
higher
LLMBoard eligible
No
Evaluation key
overall

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

FAQ

Common questions about MRCR v2 (8-needle).

Which model scores highest on MRCR v2 (8-needle)?

Muse Spark 1.3 is currently ranked first with 98.50%.

What are the top three models on MRCR v2 (8-needle)?

The current leaders are Muse Spark 1.3 (98.50%), Gemini 3.7 Flash (97.00%), and Qwen3.8 Max (92.90%).

Which MRCR v2 (8-needle) model has the lowest official input price?

GPT-5.6 Luna has the lowest matched official input price at $0.20 input / $1.2 output per 1M tokens.

Which models are fastest among MRCR v2 (8-needle) results?

The fastest matched records are Gemini 3.5 Flash (210.94 tok/s via Google), GPT-5.5 (134.94 tok/s via OpenAI), and Gemini 3 Flash (124.32 tok/s via Google).

Does the highest score result prove overall model quality?

No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.

What does MRCR v2 (8-needle) measure?

It measures models' ability to simultaneously track and reason about multiple pieces of information across extended conversations.

Is a higher score better?

Yes. Higher values rank better for this benchmark.

How many models are compared?

24 model results are currently shown.

Does this benchmark affect the overall score?

No. This benchmark is shown for reference but does not contribute to the overall score.

Rank #1Muse Spark 1.398.50%
Rank #2Gemini 3.7 Flash97.00%
Rank #3Qwen3.8 Max92.90%
Rank #4GPT-5.6 Sol91.50%

Ranking basisThis mrcr v2 (8-needle) AI model leaderboard uses descending score in the benchmark original unit. The leaderboard ranking keeps matched price and speed data separate from benchmark evidence.

  1. 01
    ME
    Muse Spark 1.3Meta
    Score
    98.50%
    Price
    $1.3 input / $4.3 output per 1M tokens
    Speed
    Up to 8.44 tok/s via Meta Model API

    Strengths

    • Ranks #1 of 24 compared models
    • 100th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures MRCR v2 (8-needle), not total model capability
  2. 02
    GO
    Gemini 3.7 FlashGoogle
    Score
    97.00%
    Price
    $0.75 input / $3.8 output per 1M tokens
    Speed
    Up to 117.86 tok/s via Google

    Strengths

    • Ranks #2 of 24 compared models
    • 96th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures MRCR v2 (8-needle), not total model capability
  3. 03
    AC
    Alibaba Cloud / Qwen Team
    Score
    92.90%
    Price
    $2.0 input / $6.0 output per 1M tokens
    Speed
    Up to 60.87 tok/s via DeepInfra

    Strengths

    • Ranks #3 of 24 compared models
    • 91th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures MRCR v2 (8-needle), not total model capability
  4. 04
    OP
    OpenAI
    Score
    91.50%
    Price
    $4.0 input / $20 output per 1M tokens
    Speed
    Up to 2.36 tok/s via OpenAI

    Strengths

    • Ranks #4 of 24 compared models
    • 87th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures MRCR v2 (8-needle), not total model capability
  5. 05
    OP
    OpenAI
    Score
    89.60%
    Price
    $2.0 input / $12 output per 1M tokens
    Speed
    Up to 99.35 tok/s via OpenAI

    Strengths

    • Ranks #5 of 24 compared models
    • 83th percentile on this benchmark
    • C evidence result

    Considerations

    • This result measures MRCR v2 (8-needle), not total model capability

Selection summary

Best AI Models for MRCR v2 (8-needle)

Muse Spark 1.3 currently leads MRCR v2 (8-needle) with 98.50%. 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.

Qwen3.8 Max
GPT-5.6 Sol
GPT-5.6 Terra
Benchmark rank #1Muse Spark 1.398.50% · $1.3 input / $4.3 output per 1M tokens
Benchmark rank #2Gemini 3.7 Flash97.00% · $0.75 input / $3.8 output per 1M tokens
Benchmark rank #3Qwen3.8 Max92.90% · $2.0 input / $6.0 output per 1M tokens