Home / Models / Z.ai / GLM 5.3 Flash

GLM 5.3 Flash pricing and routing profile

GLM 5.3 Flash from Z.ai lists at $0.15 per million input tokens and $0.50 per million output tokens, with a 1,310,720-token context window. Prices from OpenRouter's catalogue on 2026-09-28.

Input, per 1M tokens$0.15
Output, per 1M tokens$0.50
Cached input, per 1M$0.03
Context window1.31M
Confidence signalAvailable

Tool callingStructured outputReasoning controlLog probabilities

Cost per 1,000 requests

Request shapeList cost
Classification or routing (600 tokens in, 10 out)$0.095
Chat reply (2,000 in, 400 out)$0.50
Retrieval-augmented answer (8,000 in, 500 out)$1.45

Before prompt caching. Try your own shape in the cost calculator.

Can Gimbix route to it first?

Yes. 13 of 33 tracked hosts return token log probabilities for GLM 5.3 Flash, which is the confidence signal Gimbix uses to let a cheaper model answer first.

Measured by Gimbix

On 400 real BANKING77 customer messages, GLM 5.3 Flash answered 320 correctly (95% interval 75.8%–83.6%), with 1 invalid answers, at $0.0995 per 1,000 requests and a median of 1,687 ms. See all 22 models.

Who serves GLM 5.3 Flash

HostInput / output per 1MLog probabilitiesUptime, last 30 min
InferenceNet$0.045 / $0.14no99.8%
Sail Research$0.045 / $0.60no99.3%
Sail Research$0.045 / $0.60no99.4%
OpenInference$0.05 / $0.50no100.0%
Relace$0.07 / $0.50no99.9%
DeepInfra$0.075 / $0.25no99.4%
GMICloud$0.09 / $0.30no99.1%
Inceptron$0.11 / $0.45yes96.2%
Novita$0.113 / $0.375no98.1%
Decart$0.128 / $0.425no98.4%
Phala$0.128 / $0.425no99.2%
Io Net$0.135 / $0.45yes99.7%
StreamLake$0.141 / $0.47no99.2%
Modal$0.15 / $0.50yes99.6%
BaseTen$0.15 / $0.50no99.7%
Near AI$0.15 / $0.50yes99.8%
Crusoe$0.15 / $0.50no100.0%
CoreWeave$0.15 / $0.50yes99.4%
AtlasCloud$0.15 / $0.50no99.7%
Fireworks$0.15 / $0.50yes99.8%
Friendli$0.15 / $0.50no99.9%
SiliconFlow$0.15 / $0.50no99.5%
DigitalOcean$0.15 / $0.50yes100.0%
Together$0.15 / $0.50yes99.8%
Reka$0.15 / $0.50yes100.0%
Parasail$0.15 / $0.50yes99.8%
BaseTen$0.15 / $0.50no99.4%
Venice$0.15 / $0.50no99.5%
Z.AI$0.15 / $0.50no99.4%
Morph$0.162 / $0.568yes100.0%
NextBit$0.165 / $0.55no100.0%
Cloudflare$0.30 / $1.00yes99.9%
Wafer$1.00 / $0.50yes100.0%

A host that lists log probabilities may still omit them; Gimbix checks each one before relying on it.

Cheaper models that can answer first

Models priced below GLM 5.3 Flash that expose a confidence signal. Whether they answer your traffic as well is what a Gimbix pilot measures.

ModelInput / output per 1MCheaper on a chat reply
Qwen3 VL 8B Instruct$0.117 / $0.45517%
DeepSeek V4 Flash 0423$0.14 / $0.2822%
Qwen3 VL 32B Instruct$0.104 / $0.41625%
Ternary Bonsai 2 27B$0.075 / $0.5030%
Nemotron 3 Super$0.08 / $0.4532%
Gemma 3 27B$0.08 / $0.4532%

Compare GLM 5.3 Flash

Questions

How much does GLM 5.3 Flash cost?

GLM 5.3 Flash lists at $0.15 per million input tokens and $0.50 per million output tokens on OpenRouter as of 2026-09-28. A typical chat reply of 2,000 input and 400 output tokens costs about $0.50 per 1,000 requests.

What is the context window of GLM 5.3 Flash?

1,310,720 tokens.

Can a cheaper model replace GLM 5.3 Flash?

Sometimes. Gimbix proves it on your own traffic before routing: the cheaper model must match GLM 5.3 Flash on one half of your requests and make at most three answers worse on the other.