Skip to content
aiqbench

Zai · model passport

Z.ai: GLM 5.3

glm-5-3
  • Open weights
  • MIT licence
  • 753.3B params
  • 1,048,576 context
  • Released 2026-08-25(HF repo created)
11/27axes covered
Overall · LMArena
1,479#25 of 396
Best percentile
94.9Instruction following
Cohort standing
72.9median 60.2

What the data says

Benchmarked on 11 of 27 tracked axes. Every figure below is the model's own published result on that board.

Strongest axes

  • 94.9 pct — Instruction following (1,475)
  • 94.4 pct — Hard prompts (1,503)
  • 93.9 pct — Overall (1,479)

Uneven: 49.8 percentile points between its best and worst axis — 94.9 pct on Instruction following versus 45.1 pct on SimpleQA Verified.

  • Instruction following94.9 pct1,475Verify →
  • Hard prompts94.4 pct1,503Verify →
  • Overall93.9 pct1,479Verify →
  • Coding91.6 pct1,519Verify →
  • GPQA Diamond86.4 pct90.9%Verify →
  • WebDev Arena86.3 pct1,621Verify →
  • Factuality Arena82 pct1,466Verify →
  • AIME (math)73 pct91.1%Verify →
  • Terminal-Bench 4.065.2 pct41.8%Verify →
  • Agent Arena58.2 pct0.020Verify →
  • SimpleQA Verified45.1 pct41.0%Verify →
  • Vision ArenaNot benchmarked
  • Search ArenaNot benchmarked
  • Document ArenaNot benchmarked
  • FrontierMathNot benchmarked
  • ARC-AGI-2Not benchmarked
  • Humanity's Last ExamNot benchmarked
  • SimpleBenchNot benchmarked
  • LiveBenchNot benchmarked
  • BFCL v4 (function calling)Not benchmarked
  • SWE-bench Verified (live board)Not benchmarked
  • MMLU-Pro (Artificial Analysis)Not benchmarked
  • LiveCodeBench (Artificial Analysis)Not benchmarked
  • IFBench (Artificial Analysis)Not benchmarked
  • HHEM Hallucination (Vectara)Not benchmarked
  • SWE-bench VerifiedNot benchmarked
  • Terminal-BenchNot benchmarked

Percentiles are the model's standing within each board's own population. Axes marked "not benchmarked" have no published result for this model — that is an absence of data, not a low score.

How to run it

Run it yourself

Repositoryzai-org/GLM-5.3
Parameters753.3B
ArchitectureGlmMoeDsaForCausalLM
LicenceMIT

Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "zai-org/GLM-5.3"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

Ollama

We don't sync Ollama tags, so we won't print a command we can't verify — a wrong one would pull the wrong model. Search the Ollama library →

Unsloth fine-tuning recipes are model-specific and we don't derive them here. Unsloth docs →

Use it hosted

API model IDz-ai/glm-5.3
Context length1,048,576 tokens
Max output943,718 tokens
Input price$0.08 per 1M tokens
Cached input price$0.08 per 1M tokens
Output price$1.40 per 1M tokens

API parameters this endpoint supports

Frequency penaltyInclude reasoningLogit biasLogprobsMax tokensMin-pParallel Tool CallsPresence penaltyReasoningReasoning effortRepetition penaltyResponse formatSeedStop sequencesStructured outputsTemperatureTool choiceTool callingTop-kTop logprobsTop-p

Example request

curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"z-ai/glm-5.3","messages":[{"role":"user","content":"Hello"}]}'

View on OpenRouter →

Identifiers, limits and prices are synced from the upstream source shown above. Any field we haven't synced is left out entirely rather than guessed.

  • glm-5.3-max stands at 72.9 (raw mean 79.2) — measured on 11 of 27 axes.

    standingtemplateVerify →

  • glm-5.3-max stands at 72.9 versus a cohort median of 60.2 across 183 models released within 180 days — above the peer median.

    cohorttemplateVerify →

Every claim is a computed fact from the current board data, phrased by a low-temperature model call and rejected by a validation gate if any number is not in the source row. How the leaderboards work →