Skip to content
aiqbench

Zai · model passport

Z.ai: GLM 4.5 Air

glm-4-5-air
  • Open weights
  • MIT licence
  • 110.5B params
  • 131,072 context
  • Released 2025-07-28(curated)
10/27axes covered
Overall · LMArena
1,373#181 of 396
Best percentile
75.3MMLU-Pro (Artificial Analysis)
Cohort standing
51.2median 51.9

What the data says

Benchmarked on 10 of 27 tracked axes. Every figure below is the model's own published result on that board.

Uneven: 59.6 percentile points between its best and worst axis — 75.3 pct on MMLU-Pro (Artificial Analysis) versus 15.7 pct on Factuality Arena.

  • MMLU-Pro (Artificial Analysis)75.3 pct81.5%Verify →
  • LiveCodeBench (Artificial Analysis)74 pct68.4%Verify →
  • LiveBench63.9 pct60.5%Verify →
  • Coding55.8 pct1,427Verify →
  • Instruction following55.1 pct1,361Verify →
  • Hard prompts55.1 pct1,391Verify →
  • Overall54.5 pct1,373Verify →
  • IFBench (Artificial Analysis)34 pct37.6%Verify →
  • Humanity's Last Exam32.6 pct8.1%Verify →
  • Factuality Arena15.7 pct1,394Verify →
  • WebDev ArenaNot benchmarked
  • Vision ArenaNot benchmarked
  • Search ArenaNot benchmarked
  • Document ArenaNot benchmarked
  • Agent ArenaNot benchmarked
  • GPQA DiamondNot benchmarked
  • AIME (math)Not benchmarked
  • FrontierMathNot benchmarked
  • SimpleQA VerifiedNot benchmarked
  • ARC-AGI-2Not benchmarked
  • SimpleBenchNot benchmarked
  • BFCL v4 (function calling)Not benchmarked
  • SWE-bench Verified (live board)Not benchmarked
  • Terminal-Bench 4.0Not 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-4.5-Air
Parameters110.5B
ArchitectureGlm4MoeForCausalLM
LicenceMIT

Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "zai-org/GLM-4.5-Air"
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-4.5-air
Context length131,072 tokens
Max output98,304 tokens
Input price$0.13 per 1M tokens
Cached input price$0.03 per 1M tokens
Output price$0.85 per 1M tokens

API parameters this endpoint supports

Frequency penaltyInclude reasoningMax tokensPresence penaltyReasoningRepetition penaltySeedStop sequencesTemperatureTool choiceTool callingTop-kTop-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-4.5-air","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-4.5-air stands at 51.2 (raw mean 51.6) — measured on 10 of 27 axes.

    standingtemplateVerify →

  • glm-4.5-air stands at 51.2 versus a cohort median of 51.9 across 145 models released within 180 days — below 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 →