Z.ai

GLM 5.3 Flash

Launched 2026-08-26 · Listed on OpenRouter 2026-08-26

Prices updated . Benchmarks updated .

GLM-5.3-Flash is a native multimodal model from Z.ai. It is suited for efficient coding and long-horizon agent tasks. Its hybrid sparse and linear attention architecture maintains accurate long-context behavior while...

You send / 1M

$0.040

tokens in

It replies / 1M

$0.078

tokens out

Context

1M

tokens it can read at once

Typical request

$0.0001

1,000 words in, 400 out

Benchmarks

BenchmarkScoreKindDateSource
Epoch Capabilities Index153.2Independent2026-08-20Epoch AI
GPQA DiamondMax effort90.2%Independent2026-08-26Epoch AI
OTIS Mock AIMEMax effort93.9%Independent2026-08-26Epoch AI
ARC-AGI-2Max effort65.8%Independent2026-08-20ARC Prize, republished by Epoch AI
GDP.pdfMax effort14.0%Independent2026-08-20Surge AI GDP.pdf, republished by Epoch AI
CritPt15.4%Independent2026-08-20CritPt, republished by Epoch AI
Arena Elo (text)1,473Independent2026-10-28LMArena
Arena Elo (webdev)1,607Independent2026-10-28LMArena
Arena Elo (vision)1,294Independent2026-10-19LMArena

Rankings on this site use the Epoch Capabilities Index, an independent composite. Lab-reported rows are the lab's own figures and are kept separate.

Data privacy

No training or retention: Yes. At least one provider serves it with zero data retention.

ProviderYou send / 1MIt replies / 1MUptimePrivacy policy
Relace · fp4$0.040$0.078100%Read
Wafer · nvfp4$0.042$0.184100%Read
Morph · fp8$0.072$0.20898%Read
Makora · fp4$0.140$0.240100%n/a
Wafer · nvfp4$0.065$0.240100%Read
Particle AI · nvfp4$0.065$0.250100%Read
Tenstorrent · fp8$0.075$0.250100%Read
Novita · fp8$0.084$0.280100%Read
Reka$0.120$0.300100%Read
River$0.096$0.320100%Read
InferenceNet · fp4$0.108$0.350100%Read
InferenceNet · fp4$0.108$0.350100%Read
DigitalOcean$0.113$0.375100%Read
CoreWeave · nvfp4$0.120$0.380100%Read
SiliconFlow · fp8$0.100$0.40099%Read
NextBit · fp4$0.120$0.450100%Read
Inceptron · fp8$0.175$0.490100%Read
DekaLLM$0.100$0.50097%Read
Mistral$0.150$0.500100%Read
Particle AI · nvfp4$0.130$0.500100%Read
Cohere · nvfp4$0.150$0.50098%Read
Nebius · fp4$0.150$0.500100%Read
Modal · fp4$0.150$0.500100%Read
OpenInference · fp4$0.060$0.500100%Read
Near AI · fp8$0.080$0.500100%Read
Crusoe · fp4$0.150$0.500100%Read
Fireworks$0.150$0.500100%Read
Together$0.150$0.500100%Read
Parasail · fp4$0.150$0.500100%Read
BaseTen · fp8$0.150$0.500100%Read
Venice$0.150$0.500100%Read
DeepInfra · fp4$0.150$0.500100%Read
Z.AI · fp8$0.150$0.500100%Read
Parasail · fp4$0.188$0.625100%Read
Sail Research · fp4$0.175$0.625100%Read
Sail Research · fp4$0.175$0.625100%Read
Fireworks$0.225$0.75096%Read

Based on OpenRouter and Vercel AI Gateway data. Confirm terms with the provider before sending sensitive data.

What it can do

  • Reads: text, image, video, pdf
  • Writes: text
  • Weights: Open weights
  • Reasoning: Yes
  • Tool calling: Yes
  • Longest reply: 131K tokens
  • Repeated text is billed at $0.030 per 1M (prompt caching)

Compare with

Appears in

Estimate your monthly cost with GLM 5.3 Flash

Sources and licenses

  • OpenRouter: Prices, context windows, modalities and the model list come from the OpenRouter public API. Data: OpenRouter.
  • Vercel AI Gateway: The no-training flag comes from the Vercel AI Gateway model list.
  • models.dev: Lab launch dates, open-weights flags and knowledge cutoffs come from models.dev. Copyright (c) 2025 models.dev, MIT License.
  • Epoch AI: Independent benchmark scores and the Epoch Capabilities Index (ECI) come from Epoch AI's Benchmarking Hub, used under CC BY 4.0. Epoch AI, Capabilities & benchmarking, epoch.ai/benchmarks. Scores Epoch marks as external keep their original license, and the source is named on each score.
  • LMArena: Arena Elo (text, webdev and vision) comes from the LMArena leaderboard dataset (lmarena-ai/leaderboard-dataset), CC BY 4.0, leaderboard date 2026-10-28.
  • Lab-reported scores, when an editor has added them, are the lab's own numbers. Each one links to the source and is labeled lab-reported. They are not used to rank the lists.
  • What a score measures: the benchmark guide.