Home / Z.ai / GLM / GLM 5.3 Flash
GLM 5.3 Flash API price
z-ai/glm-5.3-flash
The cheapest GLM 5.3 Flash API is $0.07/M input at InferenceNet (via OpenRouter) - all 35 providers compared below, per 1M tokens.
Also known as Ox Alpha - this model ran as the anonymous OpenRouter stealth model “Ox Alpha” before Z.ai revealed it.
FlagshipZ.ai / Zhipu1.0M context35 sellers
Input
$0.07
per 1M tokens · InferenceNet (via OpenRouter)
up to $0.15
Output
$0.05
per 1M tokens · InferenceNet (via OpenRouter)
up to $1.60
Where to buy it
Cheapest GLM 5.3 Flash API providers compared - 35 sellers, from $0.07/M.
Cost calculator
| Provider | Input $/M | Output $/M | Cached in | TPS | Uptime | Your cost |
|---|---|---|---|---|---|---|
| InferenceNet (via OpenRouter)fp4 | 0.070cheapest | 0.050 | 0.030 | - | 100% | |
| StreamLake (via OpenRouter)fp8 | 0.060 | 0.200 | 0.012 | - | 100% | |
| DeepInfra (via OpenRouter)fp4 | 0.075 | 0.250 | 0.015 | - | - | |
| Novita (via OpenRouter)fp8 | 0.084 | 0.280 | 0.017 | - | 100% | |
| GMICloud (via OpenRouter)fp8 | 0.090 | 0.300 | 0.018 | - | 100% | |
| Decart (via OpenRouter)fp4 | 0.093 | 0.310 | 0.019 | - | 100% | |
| RunInfra (inference host) | 0.080 | 0.330 | 0.020 | 254 | - | |
| Near AI (via OpenRouter)fp8 | 0.105 | 0.350 | 0.025 | - | 100% | |
| Phala (via OpenRouter)nvfp4 | 0.113 | 0.375 | 0.022 | - | 100% | |
| Relace (via OpenRouter) | 0.040 | 0.500 | 0.013 | - | 100% | |
| Wafer (via OpenRouter) | 0.060 | 0.500 | 0.012 | - | 100% | |
| Sail Research (via OpenRouter) (us)fp4 | 0.045 | 0.600 | 0.029 | - | 100% | |
| Inceptron (via OpenRouter)fp8 | 0.096 | 0.550 | 0.099 | - | 100% | |
| OrcaRouter (aggregator) | 0.150 | 0.500 | - | - | - | |
| AtlasCloudfp8 | 0.150 | 0.500 | 0.030 | - | 100% | |
| BaseTenfp8 | 0.150 | 0.500 | 0.030 | - | 100% | |
| CoreWeave (via OpenRouter)nvfp4 | 0.150 | 0.500 | 0.050 | - | 100% | |
| Crusoe (via OpenRouter)fp4 | 0.150 | 0.500 | 0.030 | - | 100% | |
| DeepInfrafp4 | 0.150 | 0.500 | 0.030 | - | - | |
| DigitalOcean (via OpenRouter) | 0.150 | 0.500 | 0.030 | - | 100% | |
| Fireworks (via OpenRouter) | 0.150 | 0.500 | 0.030 | - | 100% | |
| Friendli | 0.150 | 0.500 | 0.030 | - | 100% | |
| Modal (via OpenRouter)nvfp4 | 0.150 | 0.500 | 0.030 | - | 100% | |
| Novitafp8 | 0.150 | 0.500 | 0.030 | - | 100% | |
| Parasailfp4 | 0.150 | 0.500 | 0.030 | - | 100% | |
| Phalanvfp4 | 0.150 | 0.500 | 0.030 | - | 100% | |
| SiliconFlowfp8 | 0.150 | 0.500 | 0.030 | - | 70% | |
| Together (via OpenRouter) | 0.150 | 0.500 | 0.030 | - | 100% | |
| Venice (via OpenRouter) | 0.150 | 0.500 | 0.030 | - | 100% | |
| Z.AIfp8 | 0.150 | 0.500 | 0.030 | - | 100% | |
| Together AI (inference host) | 0.150 | 0.500 | 0.030 | - | - | |
| TokenRouter (aggregator) | 0.150 | 0.500 | 0.030 | - | - | |
| Morph (via OpenRouter)fp8 | 0.060 | 1.00 | 0.015 | - | 100% | |
| OpenInference (via OpenRouter)fp4 | 0.032 | 1.20 | 0.032 | - | 100% | |
| Reka (via OpenRouter) | 0.060 | 1.60 | 0.040 | - | 100% | |
| InferX (inference host)promo, 70% off | 0.090 | 0.300 | 0.0090 | - | - |
4.7× between the cheapest and the dearest of 35 sellers.
Against its tier
- This model$0.12
- Tier median · Gemini 3.1 Pro$11.7313
- Dearest · GPT 5.5 Pro (04-23 snapshot)$210.00
#1 of 36 flagship · cheaper than 100%
Details
- Context window
- 1.0M1,048,576 tokens
- Developer
- Z.ai / Zhipu
- Intelligence
- 41.8
- Sellers
- 354.7× spread
- Cached input
- $0.03where the source publishes one
- Cheapest at
- InferenceNet (via OpenRouter)
- Combined
- $0.12in + out per 1M
- Last checked
- 2026-10-12
- Throughput
- 254.1 tok/sas published by RunInfra (inference host)
- Availability
- 100.00%median of 31 OpenRouter endpoints, last 30 min