GLM-5.3-Flash

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 reducing compute overhead.

zhipuai/glm-5.3-flash

Organization: Zhipu AI
Family: glm-flash
Providers: 82
Context: 1,000,000
Output limit: 131,072
Knowledge: —
Release: 2026-08-26
Updated: —
Weights: Open
Input: text, image/jpeg, image/png, video/mp4, video/x-matroska, video/quicktime, application/pdf
Output: text
Capabilities: —

Providers

Compare every cataloged offer for this model, including its provider, route, token limits, price, and capabilities.

141 offers · USD per million tokens

Provider	Model ID	Context	Output	Input / output · 1M	Reasoning	Tools	Structured	Details
302.AI	glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
above.dev	glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
ai&	zai-org/glm-5.3-flash	1,048,550	131,072	— / —	—	—	—	View
AIHubMix	glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
AIHubMix	coding-glm-5.3-flash-free	1,000,000	131,072	— / —	—	—	—	View
AI/ML API	z-ai/glm-5.3-flash	1,048,576	131,072	— / —	—	—	—	View
AMD	GLM-5.3-Flash	262,144	131,072	— / —	—	—	—	View
Arcee	zai-org/glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
Baseten	zai-org/GLM-5.3-Flash	1,048,576	131,072	— / —	—	—	—	View
Berget.AI	zai-org/GLM-5.3-Flash	1,048,576	16,384	— / —	—	—	—	View
Bothub	glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
Boundless	glm-5.3-flash	1,048,576	131,072	— / —	—	—	—	View
Charm Hyper	glm-5.3-flash	1,048,576	131,072	— / —	—	—	—	View
ClinePass	cline-pass/glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
Cloudflare Workers AI	@cf/zai-org/glm-5.3-flash	1,048,576	1,048,576	— / —	—	—	—	View
CommandCode	z-ai/glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
CommandCode	z-ai/glm-5.3-flashx	1,000,000	131,072	— / —	—	—	—	View
CoreWeave	zai-org/GLM-5.3-Flash	1,048,576	1,048,576	— / —	—	—	—	View
Cortecs	glm-5.3-flash	1,048,576	1,048,576	— / —	—	—	—	View
CrossModel	z-ai/glm-5.3-flash	1,000,000	128,000	— / —	—	—	—	View
Deep Infra	zai-org/GLM-5.3-Flash	1,048,576	131,072	— / —	—	—	—	View
DevPass (LLM Gateway)	glm-5.3-flash	1,048,576	131,072	— / —	—	—	—	View
DigitalOcean	glm-5.3-flash	1,048,576	131,072	— / —	—	—	—	View
Eden AI	zai/glm-5.3-flash	1,048,576	131,072	— / —	—	—	—	View
EmpirioLabs AI	glm-5-3-flash	1,000,000	131,072	— / —	—	—	—	View
engy	glm-5.3-flash	327,680	32,768	— / —	—	—	—	View
Fireworks AI	accounts/fireworks/models/glm-5p3-flash	1,048,573	131,072	— / —	—	—	—	View
Fireworks AI	accounts/fireworks/routers/glm-flash-latest	1,048,573	131,072	— / —	—	—	—	View
Friendli	zai-org/GLM-5.3-Flash	1,048,576	131,072	— / —	—	—	—	View
GreenPT	glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
Hugging Face	zai-org/GLM-5.3-Flash	1,048,576	131,072	— / —	—	—	—	View
Inceptron	zai-org/GLM-5.3-Flash	1,048,576	1,048,576	— / —	—	—	—	View
Inco	glm-5.3-flash:fast	1,000,000	131,072	— / —	—	—	—	View
IteraCompute	z-ai/glm-5.3-flash	1,048,576	131,072	— / —	—	—	—	View
Kenari	glm-5-3-flash	1,000,000	131,072	— / —	—	—	—	View
Kilo Gateway	z-ai/glm-5.3-flash	1,048,575	943,717	— / —	—	—	—	View
LLM Gateway	greenference/glm-5.3-flash	1,048,576	131,072	— / —	—	—	—	View
LLM Gateway	zai/glm-5.3-flash	1,048,576	131,072	— / —	—	—	—	View
LLM Gateway	novita/glm-5.3-flash	1,048,576	131,072	— / —	—	—	—	View
LLM Gateway	reka-ai/glm-5.3-flash	262,144	262,144	— / —	—	—	—	View
LLM Gateway	entrim/glm-5.3-flash	1,048,576	131,072	— / —	—	—	—	View
LLM Gateway	fireworks/glm-5.3-flash	1,048,576	131,072	— / —	—	—	—	View
LLM Gateway	scx-ai-gp/glm-5.3-flash	1,048,576	131,072	— / —	—	—	—	View
LLM Gateway	consensusprotocol/glm-5.3-flash	1,048,576	131,072	— / —	—	—	—	View
LLM Gateway	inference.net/glm-5.3-flash	1,048,576	128,000	— / —	—	—	—	View
LLM Gateway	runware/glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
LLM Gateway	together-ai/glm-5.3-flash	1,048,576	943,717	— / —	—	—	—	View
Melious	glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
Merge Gateway	zai/glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
Mistral	zai-glm-5-3-flash	1,048,576	131,072	— / —	—	—	—	View
Modal	nvidia/GLM-5.3-Flash-NVFP4	1,000,000	131,072	— / —	—	—	—	View
Modal	zai-org/GLM-5.3-Flash	1,000,000	131,072	— / —	—	—	—	View
NaN	glm5.3-flash	1,000,000	131,072	— / —	—	—	—	View
NanoGPT	z-ai/glm-5.3-flash	1,048,576	131,072	— / —	—	—	—	View
NanoGPT	TEE/glm-5.3-flash	1,048,576	131,072	— / —	—	—	—	View
NEAR AI Cloud	z-ai/glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
Nebius Token Factory	zai-org/GLM-5.3-Flash	1,024,000	1,024,000	— / —	—	—	—	View
Neon	glm-5-3-flash	1,048,576	131,072	— / —	—	—	—	View
Neuralwatt	glm-5.3-flash	1,048,560	1,048,560	— / —	—	—	—	View
Neuralwatt	glm-5.3-flash-flex	1,048,560	1,048,560	— / —	—	—	—	View
Novita AI	zai-org/glm-5.3-flash	1,048,576	131,072	— / —	—	—	—	View
Nvidia	z-ai/glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
Ofox	z-ai/glm-5.3-flashx	1,000,000	131,072	— / —	—	—	—	View
Ofox	z-ai/glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
Ollama Cloud	glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
OpenCode	glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
OpenCode	glm-5.3-flash-free	200,000	32,000	— / —	—	—	—	View
OpenCode Go	glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
AtlasCloudvia OpenRouter	z-ai/glm-5.3-flash fp8	1,048,576	131,072	$0.15 / $0.50	Yes	Yes	No	View
BaseTenvia OpenRouter	z-ai/glm-5.3-flash fp8	1,048,576	131,072	$0.15 / $0.50	Yes	Yes	Yes	View
CoreWeavevia OpenRouter	z-ai/glm-5.3-flash nvfp4	1,048,576	943,718	$0.15 / $0.50	Yes	Yes	Yes	View
Crusoevia OpenRouter	z-ai/glm-5.3-flash fp4	1,048,576	943,718	$0.15 / $0.50	Yes	Yes	Yes	View
Decartvia OpenRouter	z-ai/glm-5.3-flash fp4	1,048,576	943,718	$0.093 / $0.31	Yes	Yes	Yes	View
DeepInfravia OpenRouter	z-ai/glm-5.3-flash fp4	1,048,576	131,072	$0.075 / $0.25	Yes	Yes	Yes	View
DekaLLMvia OpenRouter	z-ai/glm-5.3-flash unknown	1,048,576	943,718	$0.10 / $1.00	Yes	Yes	Yes	View
DigitalOceanvia OpenRouter	z-ai/glm-5.3-flash unknown	1,048,576	131,072	$0.15 / $0.50	Yes	Yes	Yes	View
Fireworksvia OpenRouter	z-ai/glm-5.3-flash unknown	1,048,576	943,718	$0.15 / $0.50	Yes	Yes	Yes	View
Fireworksvia OpenRouter	z-ai/glm-5.3-flash unknown	1,048,576	943,718	$0.225 / $0.75	Yes	Yes	Yes	View
Friendlivia OpenRouter	z-ai/glm-5.3-flash unknown	1,048,576	943,718	$0.15 / $0.50	Yes	Yes	Yes	View
GMICloudvia OpenRouter	z-ai/glm-5.3-flash fp8	1,048,576	943,718	$0.09 / $0.30	Yes	Yes	No	View
Inceptronvia OpenRouter	z-ai/glm-5.3-flash fp8	1,048,576	943,718	$0.12 / $0.55	Yes	Yes	Yes	View
InferenceNetvia OpenRouter	z-ai/glm-5.3-flash fp4	1,048,576	262,144	$0.07 / $0.50	Yes	Yes	Yes	View
Modalvia OpenRouter	z-ai/glm-5.3-flash nvfp4	1,048,576	943,718	$0.15 / $0.50	Yes	Yes	Yes	View
Morphvia OpenRouter	z-ai/glm-5.3-flash fp8	1,048,576	943,718	$0.117 / $0.75	Yes	Yes	Yes	View
Near AIvia OpenRouter	z-ai/glm-5.3-flash fp8	1,048,576	943,718	$0.105 / $0.35	Yes	Yes	Yes	View
Novitavia OpenRouter	z-ai/glm-5.3-flash fp8	1,048,576	131,072	$0.084 / $0.28	Yes	Yes	No	View
OpenInferencevia OpenRouter	z-ai/glm-5.3-flash fp4	1,048,576	943,718	$0.044 / $0.45	Yes	Yes	Yes	View
Parasailvia OpenRouter	z-ai/glm-5.3-flash fp4	1,048,576	943,718	$0.1875 / $0.625	Yes	Yes	Yes	View
Parasailvia OpenRouter	z-ai/glm-5.3-flash fp4	1,048,576	943,718	$0.15 / $0.50	Yes	Yes	Yes	View
Phalavia OpenRouter	z-ai/glm-5.3-flash nvfp4	1,048,576	131,072	$0.1125 / $0.375	Yes	Yes	Yes	View
Rekavia OpenRouter	z-ai/glm-5.3-flash unknown	262,144	235,929	$0.06 / $1.60	Yes	Yes	Yes	View
Relacevia OpenRouter	z-ai/glm-5.3-flash unknown	1,048,576	131,072	$0.04 / $0.50	Yes	Yes	No	View
Sail Researchvia OpenRouter	z-ai/glm-5.3-flash fp4	1,048,576	131,072	$0.045 / $0.60	Yes	Yes	Yes	View
Sail Researchvia OpenRouter	z-ai/glm-5.3-flash fp4	1,048,576	131,072	$0.045 / $0.60	Yes	Yes	Yes	View
SiliconFlowvia OpenRouter	z-ai/glm-5.3-flash fp8	1,048,576	262,144	$0.15 / $0.50	Yes	Yes	No	View
StreamLakevia OpenRouter	z-ai/glm-5.3-flash fp8	1,024,000	128,000	$0.069 / $0.23	Yes	Yes	No	View
Togethervia OpenRouter	z-ai/glm-5.3-flash unknown	1,048,575	943,717	$0.15 / $0.50	Yes	Yes	Yes	View
Venicevia OpenRouter	z-ai/glm-5.3-flash unknown	1,048,576	131,072	$0.15 / $0.50	Yes	Yes	Yes	View
Wafervia OpenRouter	z-ai/glm-5.3-flash unknown	1,048,576	131,072	$0.07 / $0.50	Yes	Yes	Yes	View
Z.AIvia OpenRouter	z-ai/glm-5.3-flash fp8	1,048,576	131,072	$0.15 / $0.50	Yes	Yes	No	View
Opper	glm-5.3-flash	1,000,000	128,000	— / —	—	—	—	View
OrcaRouter	z-ai/glm-5.3-flash-free	1,000,000	128,000	— / —	—	—	—	View
OrcaRouter	z-ai/glm-5.3-flash	1,000,000	128,000	— / —	—	—	—	View
Pareto Inference	z-ai/glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
Perplexity Agent	perplexity/glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
Prime Inference	z-ai/glm-5.3-flash	1,048,576	131,072	— / —	—	—	—	View
Privatemode AI	glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
Privatemode AI	glm-flash-latest	1,000,000	131,072	— / —	—	—	—	View
Requesty	glm-5.3-flash@eu	1,000,000	262,144	— / —	—	—	—	View
Requesty	glm-5.3-flash	1,000,000	262,144	— / —	—	—	—	View
RunInfra	zai-org/GLM-5.3-Flash	1,048,576	32,768	— / —	—	—	—	View
SCNet Token Plan	GLM-5.3-Flash	1,000,000	131,072	— / —	—	—	—	View
SiliconFlow	zai-org/GLM-5.3-Flash	1,049,000	262,000	— / —	—	—	—	View
Standard Compute	z-ai/glm-5.3-flash	1,000,000	50,000	— / —	—	—	—	View
Synthetic	hf:zai-org/GLM-5.3-Flash	524,288	65,536	— / —	—	—	—	View
Tempr Gateway	zai/glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
Tempr Gateway	zai/glm-5.3-flashx	1,000,000	131,072	— / —	—	—	—	View
TensorX	z-ai/glm-5.3-flash	1,048,576	64,000	— / —	—	—	—	View
Thinking Machines	zai-org/GLM-5.3-Flash:peft:262144	262,144	262,144	— / —	—	—	—	View
Together AI	zai-org/GLM-5.3-Flash	1,048,575	400,000	— / —	—	—	—	View
TokenGo	z-ai/glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
Umans AI	umans-glm-5.3-flash	1,048,576	131,071	— / —	—	—	—	View
Umans AI	umans-coder	1,048,576	131,071	— / —	—	—	—	View
Umans AI Coding Plan	umans-coder	1,048,576	131,071	— / —	—	—	—	View
Umans AI Coding Plan	umans-glm-5.3-flash	1,048,576	131,071	— / —	—	—	—	View
Vancine	glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
Venice AI	z-ai-glm-5-3-flash	1,048,576	131,072	— / —	—	—	—	View
Vercel AI Gateway	zai/glm-5.3-flashx	1,000,000	131,072	— / —	—	—	—	View
Vercel AI Gateway	zai/glm-5.3-flash	1,000,000	131,000	— / —	—	—	—	View
Vivgrid	glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
Volcengine Ark	glm-5-3-flash-260828	1,000,000	131,072	— / —	—	—	—	View
Volcengine Ark Coding Plan	glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
Vultr	glm-5.3-flash	1,048,576	131,072	— / —	—	—	—	View
Z.AI	glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
Z.AI	glm-5.3-flashx	1,000,000	131,072	— / —	—	—	—	View
Z.AI Coding Plan	glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
ZenMux	z-ai/glm-5.3-flash	1,000,000	128,000	— / —	—	—	—	View
ZenMux	z-ai/glm-5.3-flashx	1,000,000	128,000	— / —	—	—	—	View
Zhipu AI	glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View
Zhipu AI	glm-5.3-flashx	1,000,000	131,072	— / —	—	—	—	View
Zhipu AI Coding Plan	glm-5.3-flash	1,000,000	131,072	— / —	—	—	—	View

Prices and limits apply to each offer. A dash means the value is not provided. OpenRouter offers show the hosting provider and routing channel separately.

Pricing

Compare token rates and cache charges across available routes and hosting configurations.

USD per million tokens

Input from$0.04Per 1M input tokens

Output from$0.23Per 1M output tokens

Available offers141Across 82 providers

Lowest listed input and output prices across the offers above; they may belong to different providers. These are base rates, before conditional pricing, discounts or additional fees.

Endpoint price comparisonLatest observed snapshot · 10 lowest-priced configurations

InputOutput

Relacerelace

$0.04$0.50

OpenInferenceopen-inference/fp4

$0.044$0.45

Sail Researchsail-research/fp4

$0.045$0.60

Sail Researchsail-research/us

$0.045$0.60

Rekareka

$0.06$1.60

StreamLakestreamlake/fp8

$0.069$0.23

InferenceNetinference-net/fp4

$0.07$0.50

Waferwafer

$0.07$0.50

DeepInfradeepinfra/fp4

$0.075$0.25

Novitanovita/fp8

$0.084$0.28

z-ai/glm-5.3-flashObserved Oct 10, 2026, 11:31 AM UTC

Input: $0.15
Output: $0.50
Cache read: $0.03

Hosting prices through OpenRouter All endpoints and cache rates

Endpoint	Input / 1M	Output / 1M	Cache read / 1M	Cache write / 1M
AtlasCloudatlas-cloud/fp8	$0.15	$0.50	$0.03	—
BaseTenbaseten/fp8	$0.15	$0.50	$0.03	—
CoreWeavecoreweave/nvfp4	$0.15	$0.50	$0.05	—
Crusoecrusoe/fp4	$0.15	$0.50	$0.03	—
Decartdecart/fp4	$0.093	$0.31	$0.0186	—
DeepInfradeepinfra/fp4	$0.075	$0.25	$0.015	—
DekaLLMdekallm	$0.10	$1.00	$0.04	—
DigitalOceandigitalocean	$0.15	$0.50	$0.03	—
Fireworksfireworks	$0.15	$0.50	$0.03	—
Fireworksfireworks/us	$0.225	$0.75	$0.045	—
Friendlifriendli	$0.15	$0.50	$0.03	—
GMICloudgmicloud/fp8	$0.09	$0.30	$0.018	—
Inceptroninceptron/fp8	$0.12	$0.55	$0.099	—
InferenceNetinference-net/fp4	$0.07	$0.50	$0.046	—
Modalmodal/nvfp4	$0.15	$0.50	$0.03	—
Morphmorph/fp8	$0.117	$0.75	$0.015	—
Near AInear-ai/fp8	$0.105	$0.35	$0.0245	—
Novitanovita/fp8	$0.084	$0.28	$0.0168	—
OpenInferenceopen-inference/fp4	$0.044	$0.45	$0.01	—
Parasailparasail/fast	$0.1875	$0.625	$0.0375	—
Parasailparasail/fp4	$0.15	$0.50	$0.03	—
Phalaphala/nvfp4	$0.1125	$0.375	$0.0225	—
Rekareka	$0.06	$1.60	$0.04	—
Relacerelace	$0.04	$0.50	$0.0125	—
Sail Researchsail-research/fp4	$0.045	$0.60	$0.0285	—
Sail Researchsail-research/us	$0.045	$0.60	$0.0285	—
SiliconFlowsiliconflow/fp8	$0.15	$0.50	$0.03	—
StreamLakestreamlake/fp8	$0.069	$0.23	$0.0138	—
Togethertogether	$0.15	$0.50	$0.03	—
Venicevenice	$0.15	$0.50	$0.03	—
Waferwafer	$0.07	$0.50	$0.03	—
Z.AIz-ai/fp8	$0.15	$0.50	$0.03	—

Each row is a hosting configuration. Open an offer in Providers for its full pricing conditions and additional fees. Endpoint observations may differ from the routing catalog's base price.

Performance

Review reported latency and output-throughput percentiles from the latest observed rolling window.

Latest 30-minute observation

Time to first token · milliseconds

Latency distributionRolling 30-minute snapshot · 6 endpoints

P50P75P90P99

AtlasCloudatlas-cloud/fp8

2,782.5–17,459.75 ms

CoreWeavecoreweave/nvfp4

699–6,502.62 ms

Crusoecrusoe/fp4

864.5–3,567.25 ms

Decartdecart/fp4

1,282–7,463.26 ms

DeepInfradeepinfra/fp4

1,552–7,681.12 ms

DekaLLMdekallm

609–3,758.6 ms

All latency percentiles 31 endpoints

Endpoint	P50	P75	P90	P99
AtlasCloudatlas-cloud/fp8	2,782.5 ms	6,199.5 ms	8,898.2 ms	17,459.75 ms
CoreWeavecoreweave/nvfp4	699 ms	1,053.25 ms	1,778.8 ms	6,502.62 ms
Crusoecrusoe/fp4	864.5 ms	1,109 ms	1,571.9 ms	3,567.25 ms
Decartdecart/fp4	1,282 ms	1,655 ms	2,198.9 ms	7,463.26 ms
DeepInfradeepinfra/fp4	1,552 ms	2,157 ms	3,022 ms	7,681.12 ms
DekaLLMdekallm	609 ms	928 ms	1,485.6 ms	3,758.6 ms
DigitalOceandigitalocean	575 ms	895 ms	1,329.8 ms	3,746.71 ms
Fireworksfireworks	1,046 ms	1,901 ms	3,138 ms	8,955.86 ms
Fireworksfireworks/us	625 ms	814 ms	1,054.4 ms	7,746.78 ms
Friendlifriendli	956 ms	3,025 ms	5,335.6 ms	11,259.94 ms
GMICloudgmicloud/fp8	3,351.5 ms	5,372 ms	7,844 ms	15,982.65 ms
Inceptroninceptron/fp8	538 ms	947 ms	1,877.4 ms	13,959.84 ms
InferenceNetinference-net/fp4	666 ms	1,203.25 ms	2,061.8 ms	8,137.18 ms
Modalmodal/nvfp4	289 ms	429 ms	715 ms	2,280.42 ms
Morphmorph/fp8	876 ms	1,784 ms	6,028 ms	17,930 ms
Near AInear-ai/fp8	1,196 ms	1,938 ms	4,208 ms	21,343.4 ms
Novitanovita/fp8	3,400 ms	5,358.25 ms	7,973.5 ms	22,116.23 ms
OpenInferenceopen-inference/fp4	3,844 ms	10,565.25 ms	12,006.8 ms	14,003.17 ms
Parasailparasail/fast	701 ms	1,489 ms	2,954.9 ms	10,616.64 ms
Parasailparasail/fp4	894 ms	1,690 ms	3,215 ms	9,981.89 ms
Phalaphala/nvfp4	4,197 ms	5,144 ms	7,191.5 ms	13,444.19 ms
Rekareka	883 ms	1,625.25 ms	3,288.7 ms	9,710.06 ms
Relacerelace	1,148 ms	2,009.75 ms	3,409.9 ms	14,392.42 ms
Sail Researchsail-research/fp4	638.5 ms	1,157 ms	1,987 ms	4,776 ms
Sail Researchsail-research/us	391 ms	787 ms	1,345 ms	3,472.9 ms
SiliconFlowsiliconflow/fp8	5,986.5 ms	8,968.25 ms	14,338 ms	26,991.82 ms
StreamLakestreamlake/fp8	1,244.5 ms	2,166.5 ms	3,776.1 ms	10,480.92 ms
Togethertogether	529 ms	1,056.25 ms	2,496.3 ms	12,932.8 ms
Venicevenice	1,022 ms	1,828 ms	3,067 ms	7,886.4 ms
Waferwafer	620 ms	1,240 ms	2,484.9 ms	6,927.18 ms
Z.AIz-ai/fp8	3,046 ms	5,016 ms	7,847 ms	21,320.3 ms

Output throughput · tokens per second

Throughput distributionRolling 30-minute snapshot · 6 endpoints

P50P75P90P99

AtlasCloudatlas-cloud/fp8

45–97.5 t/s

CoreWeavecoreweave/nvfp4

156–435.09 t/s

Crusoecrusoe/fp4

86–157 t/s

Decartdecart/fp4

61–90 t/s

DeepInfradeepinfra/fp4

27–52 t/s

DekaLLMdekallm

124–360.87 t/s

All throughput percentiles 31 endpoints

Endpoint	P50	P75	P90	P99
AtlasCloudatlas-cloud/fp8	45 t/s	61 t/s	70 t/s	97.5 t/s
CoreWeavecoreweave/nvfp4	156 t/s	228 t/s	329 t/s	435.09 t/s
Crusoecrusoe/fp4	86 t/s	102 t/s	119 t/s	157 t/s
Decartdecart/fp4	61 t/s	69 t/s	77 t/s	90 t/s
DeepInfradeepinfra/fp4	27 t/s	32 t/s	39 t/s	52 t/s
DekaLLMdekallm	124 t/s	177 t/s	244 t/s	360.87 t/s
DigitalOceandigitalocean	47 t/s	50 t/s	53 t/s	58.52 t/s
Fireworksfireworks	96 t/s	131 t/s	166 t/s	220 t/s
Fireworksfireworks/us	105 t/s	122.5 t/s	138 t/s	191.9 t/s
Friendlifriendli	73 t/s	153 t/s	213 t/s	324.09 t/s
GMICloudgmicloud/fp8	42 t/s	64 t/s	87 t/s	112 t/s
Inceptroninceptron/fp8	114 t/s	155 t/s	201 t/s	304 t/s
InferenceNetinference-net/fp4	105 t/s	136 t/s	163 t/s	233.09 t/s
Modalmodal/nvfp4	104 t/s	133 t/s	160 t/s	212 t/s
Morphmorph/fp8	63 t/s	106 t/s	148 t/s	217 t/s
Near AInear-ai/fp8	44 t/s	54 t/s	63 t/s	79 t/s
Novitanovita/fp8	34 t/s	46 t/s	63 t/s	103 t/s
OpenInferenceopen-inference/fp4	23 t/s	29 t/s	34 t/s	39 t/s
Parasailparasail/fast	114 t/s	161 t/s	220.7 t/s	383.54 t/s
Parasailparasail/fp4	113 t/s	160 t/s	201 t/s	278 t/s
Phalaphala/nvfp4	25 t/s	40 t/s	50 t/s	78 t/s
Rekareka	101 t/s	169 t/s	218 t/s	294 t/s
Relacerelace	63 t/s	92 t/s	129 t/s	211 t/s
Sail Researchsail-research/fp4	141 t/s	190 t/s	237 t/s	355.18 t/s
Sail Researchsail-research/us	159 t/s	205 t/s	253 t/s	345.09 t/s
SiliconFlowsiliconflow/fp8	26 t/s	41 t/s	62 t/s	119 t/s
StreamLakestreamlake/fp8	43 t/s	51 t/s	58 t/s	74.09 t/s
Togethertogether	117 t/s	146 t/s	176 t/s	261.09 t/s
Venicevenice	98 t/s	127 t/s	147 t/s	203 t/s
Waferwafer	50 t/s	59 t/s	67 t/s	89 t/s
Z.AIz-ai/fp8	42 t/s	52 t/s	90 t/s	122 t/s

Percentiles describe the observed request distribution. These are snapshots of a rolling window, not a historical time series.

Uptime

Compare successful request rates across the latest 5-minute, 30-minute, and 24-hour observations.

Successful requests by window

Availability by endpointCurrent overlapping windows · 32 endpoints

Endpoint5 minutes30 minutes24 hours

AtlasCloudatlas-cloud/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

89.47%

95.35%

BaseTenbaseten/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

—

CoreWeavecoreweave/nvfp4Observed Oct 10, 2026, 11:31 AM UTC

100%

Crusoecrusoe/fp4Observed Oct 10, 2026, 11:31 AM UTC

100%

98.1%

Decartdecart/fp4Observed Oct 10, 2026, 11:31 AM UTC

100%

DeepInfradeepinfra/fp4Observed Oct 10, 2026, 11:31 AM UTC

100%

98.75%

DekaLLMdekallmObserved Oct 10, 2026, 11:31 AM UTC

100%

99.86%

DigitalOceandigitaloceanObserved Oct 10, 2026, 11:31 AM UTC

100%

FireworksfireworksObserved Oct 10, 2026, 11:31 AM UTC

100%

99.1%

Fireworksfireworks/usObserved Oct 10, 2026, 11:31 AM UTC

—

100%

99.85%

FriendlifriendliObserved Oct 10, 2026, 11:31 AM UTC

100%

GMICloudgmicloud/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

99.86%

Inceptroninceptron/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

99.15%

InferenceNetinference-net/fp4Observed Oct 10, 2026, 11:31 AM UTC

100%

Modalmodal/nvfp4Observed Oct 10, 2026, 11:31 AM UTC

100%

Morphmorph/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

96.54%

Near AInear-ai/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

Novitanovita/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

99.86%

OpenInferenceopen-inference/fp4Observed Oct 10, 2026, 11:31 AM UTC

100%

Parasailparasail/fastObserved Oct 10, 2026, 11:31 AM UTC

100%

99.93%

Parasailparasail/fp4Observed Oct 10, 2026, 11:31 AM UTC

100%

Phalaphala/nvfp4Observed Oct 10, 2026, 11:31 AM UTC

100%

RekarekaObserved Oct 10, 2026, 11:31 AM UTC

100%

98.29%

RelacerelaceObserved Oct 10, 2026, 11:31 AM UTC

100%

Sail Researchsail-research/fp4Observed Oct 10, 2026, 11:31 AM UTC

100%

99.86%

Sail Researchsail-research/usObserved Oct 10, 2026, 11:31 AM UTC

100%

99.86%

SiliconFlowsiliconflow/fp8Observed Oct 10, 2026, 11:31 AM UTC

80%

92.86%

91.01%

StreamLakestreamlake/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

99.72%

TogethertogetherObserved Oct 10, 2026, 11:31 AM UTC

100%

VeniceveniceObserved Oct 10, 2026, 11:31 AM UTC

100%

98.26%

WaferwaferObserved Oct 10, 2026, 11:31 AM UTC

100%

Z.AIz-ai/fp8Observed Oct 10, 2026, 11:31 AM UTC

100%

Reported by OpenRouter at the time of collection; rate-limited requests are excluded. A dash means no measurement was reported. This is not a live availability check.

Benchmarks

Review attributed evaluation results on each source's original scale.

External evaluations

z-ai/glm-5.3-flashObserved Oct 10, 2026, 11:31 AM UTC

Benchmark profileArtificial Analysis · shared source scale from 0 to 80

Coding

71.5

Agentic

50.9

Intelligence

41.8

Design Arena	Category	Elo	Win rate	Rank
models	3d	1,323	56.5%	11
models	asciiart	1,272	54.9%	12
models	codecategories	1,287	49.4%	17
models	dataviz	1,266	50.3%	27
models	gamedev	1,289	47.3%	19
models	svg	1,289	57.1%	10
models	uicomponent	1,324	55.1%	9
models	website	1,276	48.4%	24

Sources: Artificial Analysis and Design Arena, via OpenRouter. Scores retain their original scales and are not OpenAI Suite ratings.

Detailed evaluations

Evaluation scoresAccuracy or item F1 · higher is better

gpqa diamondGLM 5.3 Flash

90.909%

tau bench verified airlineGLM 5.3 Flash

70%

tau bench verified airlineZ.ai: GLM 5.3 Flash

75.905%

gpqa diamondZ.ai: GLM 5.3 Flash

84.617%

Evaluation	Configuration	Source	Results	Details
gpqa diamond	GLM 5.3 Flashz-ai/glm-5.3-flash	OpenRouter	90.909%	Details
tau bench verified airline	GLM 5.3 Flashz-ai/glm-5.3-flash	OpenRouter	70%	Details
models · asciiart	GLM-5.3-Flashz-ai/glm-5.3-flash-20260826	Design Arena	1,272 Elo	Details
models · uicomponent	GLM-5.3-Flashz-ai/glm-5.3-flash-20260826	Design Arena	1,311 Elo	Details
models · codecategories	GLM-5.3-Flashz-ai/glm-5.3-flash-20260826	Design Arena	1,281 Elo	Details
models · gamedev	GLM-5.3-Flashz-ai/glm-5.3-flash-20260826	Design Arena	1,279 Elo	Details
tau bench verified airline	Z.ai: GLM 5.3 Flashz-ai/glm-5.3-flash-20260826	OpenRouter	75.905%	Details
Composite indices	GLM-5.3-Flashz-ai/glm-5.3-flash-20260826	Artificial Analysis	Intelligence 41.8 · Coding 71.5 · Agentic 50.9	Details
models · svg	GLM-5.3-Flashz-ai/glm-5.3-flash-20260826	Design Arena	1,283 Elo	Details
gpqa diamond	Z.ai: GLM 5.3 Flashz-ai/glm-5.3-flash-20260826	OpenRouter	84.617%	Details
models · 3d	GLM-5.3-Flashz-ai/glm-5.3-flash-20260826	Design Arena	1,317 Elo	Details
models · dataviz	GLM-5.3-Flashz-ai/glm-5.3-flash-20260826	Design Arena	1,262 Elo	Details
models · website	GLM-5.3-Flashz-ai/glm-5.3-flash-20260826	Design Arena	1,271 Elo	Details

Sources: Artificial Analysis, Design Arena and OpenRouter. Evaluation configurations and scales differ; results are not interchangeable. Updated Oct 24, 2026.

Apps & session costs

Compare observed median session costs for applications using this model, grouped by conversation length.

30-day sample · through Oct 18, 2026

Application	Model route	Turns	Median cost / session	Details
Kilo Code	z-ai/glm-5.3-flash-20260826	10–49 turns	$0.053504	Details
Hermes Agent	z-ai/glm-5.3-flash-20260826	50+ turns	$0.281672	Details
Hermes Agent	z-ai/glm-5.3-flash-20260826	1 turn	$0.000769	Details
Hermes Agent	z-ai/glm-5.3-flash-20260826	2–9 turns	$0.005349	Details
Kilo Code	z-ai/glm-5.3-flash-20260826	1 turn	$0.000925	Details
Kilo Code	z-ai/glm-5.3-flash-20260826	2–9 turns	$0.010016	Details
Claude Code	z-ai/glm-5.3-flash-20260826	1 turn	$0.002057	Details
Claude Code	z-ai/glm-5.3-flash-20260826	50+ turns	$0.404733	Details
Codex	z-ai/glm-5.3-flash-20260826	1 turn	$0.000677	Details
Codex	z-ai/glm-5.3-flash-20260826	50+ turns	$0.070536	Details
Codex	z-ai/glm-5.3-flash-20260826	2–9 turns	$0.005007	Details
Hermes Agent	z-ai/glm-5.3-flash-20260826	10–49 turns	$0.039417	Details
Claude Code	z-ai/glm-5.3-flash-20260826	10–49 turns	$0.05174	Details
Codex	z-ai/glm-5.3-flash-20260826	10–49 turns	$0.030582	Details
Claude Code	z-ai/glm-5.3-flash-20260826	2–9 turns	$0.007865	Details
Kilo Code	z-ai/glm-5.3-flash-20260826	50+ turns	$0.418473	Details

Source: OpenRouter, as of Oct 20, 2026. Published under CC BY 4.0. These are observed costs per session, not token prices or a forecast for your workload. Apps are compared separately.

Activity

Daily prompt and completion tokens reported for this model's routes in OpenRouter's top 50. Missing days have no published value.

Last 90 complete UTC days

Daily token usage103,932,445,028,822 reported tokens · 59 days with observations

4.4T0

2026-08-26
360,760,043,126 tokens

2026-08-27
1,380,964,831,877 tokens

2026-08-28
1,556,774,182,660 tokens

2026-08-29
1,322,948,887,485 tokens

2026-08-30
1,536,762,258,643 tokens

2026-08-31
1,979,092,565,477 tokens

2026-09-01
1,872,948,241,668 tokens

2026-09-02
1,791,420,461,439 tokens

2026-09-03
1,888,512,106,663 tokens

2026-09-04
1,922,687,542,042 tokens

2026-09-05
1,466,818,656,113 tokens

2026-09-06
1,472,192,943,090 tokens

2026-09-07
1,821,445,258,438 tokens

2026-09-08
1,808,250,248,971 tokens

2026-09-09
1,951,528,306,504 tokens

2026-09-10
1,821,255,463,174 tokens

2026-09-11
1,735,864,603,429 tokens

2026-09-12
1,343,229,501,205 tokens

2026-09-13
1,408,200,876,929 tokens

2026-09-14
1,577,050,969,622 tokens

2026-09-15
1,585,636,122,986 tokens

2026-09-16
1,798,851,266,835 tokens

2026-09-17
1,964,603,439,979 tokens

2026-09-18
1,944,494,628,152 tokens

2026-09-19
2,685,459,930,903 tokens

2026-09-20
2,515,123,497,699 tokens

2026-09-21
4,362,717,938,367 tokens

2026-09-22
3,153,115,211,486 tokens

2026-09-23
2,409,694,696,523 tokens

2026-09-24
1,964,153,145,155 tokens

2026-09-25
2,126,728,706,421 tokens

2026-09-26
1,282,011,293,247 tokens

2026-09-27
1,018,265,702,294 tokens

2026-09-28
1,317,250,703,656 tokens

2026-09-29
1,503,521,835,069 tokens

2026-09-30
1,421,695,513,697 tokens

2026-10-01
1,479,789,246,190 tokens

2026-10-02
1,650,868,242,738 tokens

2026-10-03
1,174,693,172,798 tokens

2026-10-04
1,185,380,576,794 tokens

2026-10-05
1,603,476,115,426 tokens

2026-10-06
1,673,194,955,545 tokens

2026-10-07
1,724,149,948,224 tokens

2026-10-08
1,829,336,671,487 tokens

2026-10-09
2,008,077,801,909 tokens

2026-10-10
1,604,367,743,946 tokens

2026-10-11
1,793,004,227,955 tokens

2026-10-12
2,033,311,942,256 tokens

2026-10-13
1,874,361,498,272 tokens

2026-10-14
1,797,296,946,947 tokens

2026-10-15
1,835,073,401,994 tokens

2026-10-16
1,737,946,577,166 tokens

2026-10-17
1,526,375,287,786 tokens

2026-10-18
1,426,914,694,347 tokens

2026-10-19
1,742,144,787,145 tokens

2026-10-20
1,527,645,990,143 tokens

2026-10-21
1,541,090,577,880 tokens

2026-10-22
1,913,901,671,523 tokens

2026-10-23
2,178,011,369,327 tokens

Jul 26, 2026Oct 23, 2026

Daily observations · 59 days

Date (UTC)	Reported tokens	Reported routes
Oct 23, 2026	2,178,011,369,327	1
Oct 22, 2026	1,913,901,671,523	1
Oct 21, 2026	1,541,090,577,880	1
Oct 20, 2026	1,527,645,990,143	1
Oct 19, 2026	1,742,144,787,145	1
Oct 18, 2026	1,426,914,694,347	1
Oct 17, 2026	1,526,375,287,786	1
Oct 16, 2026	1,737,946,577,166	1
Oct 15, 2026	1,835,073,401,994	1
Oct 14, 2026	1,797,296,946,947	1
Oct 13, 2026	1,874,361,498,272	1
Oct 12, 2026	2,033,311,942,256	1
Oct 11, 2026	1,793,004,227,955	1
Oct 10, 2026	1,604,367,743,946	1
Oct 9, 2026	2,008,077,801,909	1
Oct 8, 2026	1,829,336,671,487	1
Oct 7, 2026	1,724,149,948,224	1
Oct 6, 2026	1,673,194,955,545	1
Oct 5, 2026	1,603,476,115,426	1
Oct 4, 2026	1,185,380,576,794	1
Oct 3, 2026	1,174,693,172,798	1
Oct 2, 2026	1,650,868,242,738	1
Oct 1, 2026	1,479,789,246,190	1
Sep 30, 2026	1,421,695,513,697	1
Sep 29, 2026	1,503,521,835,069	1
Sep 28, 2026	1,317,250,703,656	1
Sep 27, 2026	1,018,265,702,294	1
Sep 26, 2026	1,282,011,293,247	1
Sep 25, 2026	2,126,728,706,421	1
Sep 24, 2026	1,964,153,145,155	1
Sep 23, 2026	2,409,694,696,523	1
Sep 22, 2026	3,153,115,211,486	1
Sep 21, 2026	4,362,717,938,367	1
Sep 20, 2026	2,515,123,497,699	1
Sep 19, 2026	2,685,459,930,903	1
Sep 18, 2026	1,944,494,628,152	1
Sep 17, 2026	1,964,603,439,979	1
Sep 16, 2026	1,798,851,266,835	1
Sep 15, 2026	1,585,636,122,986	1
Sep 14, 2026	1,577,050,969,622	1
Sep 13, 2026	1,408,200,876,929	1
Sep 12, 2026	1,343,229,501,205	1
Sep 11, 2026	1,735,864,603,429	1
Sep 10, 2026	1,821,255,463,174	1
Sep 9, 2026	1,951,528,306,504	1
Sep 8, 2026	1,808,250,248,971	1
Sep 7, 2026	1,821,445,258,438	1
Sep 6, 2026	1,472,192,943,090	1
Sep 5, 2026	1,466,818,656,113	1
Sep 4, 2026	1,922,687,542,042	1
Sep 3, 2026	1,888,512,106,663	1
Sep 2, 2026	1,791,420,461,439	1
Sep 1, 2026	1,872,948,241,668	1
Aug 31, 2026	1,979,092,565,477	1
Aug 30, 2026	1,536,762,258,643	1
Aug 29, 2026	1,322,948,887,485	1
Aug 28, 2026	1,556,774,182,660	1
Aug 27, 2026	1,380,964,831,877	1
Aug 26, 2026	360,760,043,126	1

Source: OpenRouter (openrouter.ai/rankings), as of Oct 24, 2026. CC BY 4.0. Totals include only individually published routes. Tokenizers differ by provider; the aggregated “other” category is never assigned to a model.

Technical details & API

Explore specifications, supported parameters, reasoning controls, and API identifiers for each OpenRouter route.

Current route specifications

z-ai/glm-5.3-flashObserved Oct 10, 2026, 11:31 AM UTC

Context: 1,048,576
Maximum output: 943,717
Tokenizer: Other
Input: text, image, video
Output: text
Moderation: No
Added to OpenRouter: 2026-08-26

Base URLhttps://openrouter.ai/api/v1

Model IDz-ai/glm-5.3-flash

Version IDz-ai/glm-5.3-flash-20260826

Weights IDzai-org/GLM-5.3-Flash

Reasoning

Required: Yes
Enabled by default: Yes
Default effort: max
Supported efforts: max, high, low

Supported parameters

frequency_penalty
include_reasoning
logit_bias
logprobs
max_tokens
min_p
parallel_tool_calls
presence_penalty
reasoning
reasoning_effort
repetition_penalty
response_format
seed
stop
structured_outputs
temperature
tool_choice
tools
top_k
top_logprobs
top_p

Default parameters

top p: 0.95
temperature: 1

Frequently asked questions

Answers based on the model metadata and serving offers currently in the catalog.

How much does GLM-5.3-Flash cost?

Listed rates start at $0.04 per million input tokens and $0.23 per million output tokens. Prices vary by provider and configuration; see Pricing for details.

Which providers offer GLM-5.3-Flash?

There are 141 cataloged offers across 82 providers. Compare model IDs, prices and limits in Providers.

What is the context window?

The cataloged model context is 1,000,000 tokens. Each serving endpoint may apply a different limit.

Does it support tools and structured output?

Tool calling: Not reported. Structured output: Not reported. Support can vary by endpoint.

Resources

Hugging Face ↗

Catalog history · 4 recorded revisions

Metadata observations since this model was first cataloged. These are not model release versions.

Oct 20, 2026 · e474bf4ecb40
Sep 16, 2026 · c51cc41cc7b0
Sep 14, 2026 · 051fd4dcc9dc
Sep 12, 2026 · 5834840b0b36