Translate
Independent model intelligence

Find the right AI model,
not just the loudest one.

Compare current model capability, context, modalities, public preference and API economics in one continuously refreshed decision surface.

LiveBench 2.0 LMArena models.dev OpenRouter AIHOT cross-check
Back to leaderboard
Zhipu AI · Active

glm-5.3-flash

glm-flash family · launched 2026-08-26

MultimodalReasoningToolsStructuredOpen weights
AITopic score74.9#18 overall

Native multimodal GLM model for efficient coding and long-horizon agent tasks

Evidence coverage95%5 evaluations · 2 institutions
Context / maximum output1M / 128ktokens
Cache / input / output$0.0282 · $0.11 · $0.39USD per million tokens
Value / freshness92.2 / 86relative value · release freshness
Product profile

What this model supports

Inputtext, image, video
Outputtext
Knowledge cutoffNot disclosed
WeightsOpen weights
Capability profile

Strengths by domain

Coding80.61 public results
Research & reasoning77.61 public results
Professional work66.51 public results
Knowledge & instruction70.11 public results
Behind the score

Every result has a source

Raw results and standardized readings are shown together. Missing tests add neither points nor penalties.

Coding1 results

LiveBench 2.0 · CodingLiveBench80.60 scoreReading 80.6

Imported from LiveBench 2.0. The published 0–100 score is used directly; it is not z-score normalized.

Configuration: LiveBench 2.0Open evidence

Research & reasoning1 results

LiveBench 2.0 · ResearchLiveBench77.60 scoreReading 77.6

Imported from LiveBench 2.0. The published 0–100 score is used directly; it is not z-score normalized.

Configuration: LiveBench 2.0Open evidence

Professional work1 results

LiveBench 2.0 · ProfessionalLiveBench66.50 scoreReading 66.5

Imported from LiveBench 2.0. The published 0–100 score is used directly; it is not z-score normalized. Professional = long-agentic, browser, data-analysis and design average.

Configuration: LiveBench 2.0Open evidence

Knowledge & instruction1 results

LiveBench 2.0 · KnowledgeLiveBench70.10 scoreReading 70.1

Imported from LiveBench 2.0. The published 0–100 score is used directly; it is not z-score normalized. Knowledge uses the published LiveBench research suite.

Configuration: LiveBench 2.0Open evidence

Human preference & indices3 results

Arena TextLMArena1,470.5 EloReading 74.9

Direct LMArena text leaderboard result.

Open evidence
LiveBenchLiveBench70.61 scoreReading 70.6

Imported from LiveBench 2.0. The published 0–100 score is used directly; it is not z-score normalized.

Configuration: LiveBench 2.0Open evidence
AA IndexArtificial Analysis67.20 indexReading 62.8

Imported from Artificial Analysis. The published 0–100 score is used directly; it is not z-score normalized.

Configuration: LiveBench 2.0Open evidence
Same builder

More from Zhipu AI