# Cheapest GLM 5.3 API providers, compared (prices checked Nov 7, 2026)

URL: https://ernd.ai/inference-prices/glm-5-3

> ernd partners with [Wafer](https://ernd.ai/partners/wafer). Prices are from OpenRouter's public listings.

GLM 5.3 is an open-weight model from Z.ai. On OpenRouter, 47 listings from 38 providers sell it, and the blended price runs from $0.402 to $4.30 per million tokens.

| Provider | Input /1M | Output /1M | Blended 3:1 | Cache read /1M | Context | Quantization | Latency | Throughput | With ernd card |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Wafer (US), ernd partner | $0.0699 | $1.40 | $0.402 | $0.0689 | 1.05M | FP4 | 0.42 s | 89 tok/s | $0.362 after 10% cash back |
| Wafer, ernd partner | $0.179 | $1.40 | $0.484 | $0.178 | 1.05M | FP4 | 0.35 s | 93 tok/s | $0.436 after 10% cash back |
| Reka AI | $0.14 | $2.10 | $0.63 | $0.105 | 1.05M | Not listed | 0.87 s | 43 tok/s | – |
| Decart | $0.56 | $1.78 | $0.865 | $0.10 | 1.05M | FP4 | 0.84 s | 92 tok/s | – |
| DekaLLM | $0.18 | $2.95 | $0.873 | $0.15 | 1.05M | Not listed | 1.26 s | 49 tok/s | – |
| inference.net | $0.08 | $3.60 | $0.96 | $0.076 | 1.05M | FP4 | 0.50 s | 98 tok/s | – |
| inference.net (US) | $0.08 | $3.60 | $0.96 | $0.076 | 1.05M | FP4 | 0.54 s | 94 tok/s | – |
| NovitaAI | $0.63 | $1.98 | $0.968 | $0.117 | 1.05M | FP8 | 1.97 s | 53 tok/s | – |
| Makora | $0.30 | $3.00 | $0.975 | $0.125 | 1.05M | FP4 | 0.77 s | 63 tok/s | – |
| DeepInfra | $0.563 | $2.50 | $1.05 | $0.125 | 1.05M | FP4 | 1.45 s | 46 tok/s | – |
| Cloudflare | $0.70 | $2.20 | $1.07 | $0.13 | 1.05M | Not listed | 1.81 s | 65 tok/s | – |
| GMICloud | $0.70 | $2.20 | $1.07 | $0.13 | 1.05M | FP8 | 2.67 s | 50 tok/s | – |
| SiliconFlow | $0.70 | $2.20 | $1.07 | $0.13 | 1.05M | FP8 | 2.04 s | 59 tok/s | – |
| Tenstorrent | $0.60 | $2.50 | $1.07 | $0.10 | 1.04M | FP8 | 0.77 s | 75 tok/s | – |
| Prime Intellect | $0.50 | $3.08 | $1.15 | $0.093 | 1.05M | Not listed | 0.85 s | 83 tok/s | – |
| Relace | $0.0703 | $4.38 | $1.15 | $0.0703 | 1.05M | FP4 | 1.03 s | 76 tok/s | – |
| AkashML | $0.19 | $4.40 | $1.24 | $0.19 | 1.05M | FP4 | 1.20 s | 87 tok/s | – |
| Phala | $0.84 | $2.64 | $1.29 | $0.156 | 1.05M | Not listed | 3.24 s | 38 tok/s | – |
| Morph | $0.74 | $3.00 | $1.30 | $0.25 | 1.05M | FP8 | 1.55 s | 32 tok/s | – |
| CoreWeave | $0.75 | $3.00 | $1.31 | $0.19 | 1.05M | NVFP4 | 0.59 s | 92 tok/s | – |
| Sail Research (US) | $0.75 | $3.00 | $1.31 | $0.12 | 1.05M | FP8 | 0.98 s | 72 tok/s | – |
| DigitalOcean | $0.91 | $2.86 | $1.40 | $0.169 | 1.05M | Not listed | 0.55 s | 88 tok/s | – |
| Fireworks Spot | $0.98 | $3.08 | $1.50 | $0.182 | 1.05M | Not listed | 0.65 s | 88 tok/s | – |
| Alibaba Cloud Int. | $1.19 | $3.74 | $1.83 | $0.238 | 1M | Not listed | 1.90 s | 52 tok/s | – |
| Sail Research | $1.05 | $4.20 | $1.84 | $0.168 | 1.05M | FP8 | 0.78 s | 73 tok/s | – |
| Friendli | $1.26 | $3.96 | $1.94 | $0.234 | 1.05M | Not listed | 0.36 s | 83 tok/s | – |
| Inceptron | $1.00 | $4.90 | $1.98 | $0.14 | 1.05M | FP4 | 0.73 s | 100 tok/s | – |
| Baidu Qianfan | $1.40 | $4.40 | $2.15 | $0.26 | 1.05M | FP8 | 1.06 s | 61 tok/s | – |
| AtlasCloud | $1.40 | $4.40 | $2.15 | $0.26 | 1.05M | FP8 | 2.83 s | 40 tok/s | – |
| Baseten (fp4, US) | $1.40 | $4.40 | $2.15 | $0.14 | 1.05M | FP4 | 2.35 s | 58 tok/s | – |
| Baseten (fp4) | $1.40 | $4.40 | $2.15 | $0.14 | 1.05M | FP4 | 2.74 s | 48 tok/s | – |
| Crusoe | $1.40 | $4.40 | $2.15 | $0.26 | 1.05M | FP4 | 0.47 s | 118 tok/s | – |
| Fireworks | $1.40 | $4.40 | $2.15 | $0.26 | 1.05M | Not listed | 1.31 s | 77 tok/s | – |
| Mistral (ZDR) | $1.40 | $4.40 | $2.15 | $0.14 | 1.05M | NVFP4 | 0.94 s | 102 tok/s | – |
| Modal | $1.40 | $4.40 | $2.15 | $0.26 | 1.05M | Not listed | 0.34 s | 112 tok/s | – |
| Nebius Token Factory | $1.40 | $4.40 | $2.15 | $0.26 | 1.05M | FP4 | 0.69 s | 87 tok/s | – |
| Parasail | $1.40 | $4.40 | $2.15 | $0.26 | 1.05M | FP8 | 0.66 s | 99 tok/s | – |
| Together | $1.40 | $4.40 | $2.15 | $0.26 | 1.05M | Not listed | 0.55 s | 101 tok/s | – |
| Venice | $1.40 | $4.40 | $2.15 | $0.26 | 1M | Not listed | 1.03 s | 72 tok/s | – |
| Z.ai | $1.40 | $4.40 | $2.15 | $0.26 | 1.05M | FP8 | 1.71 s | 53 tok/s | – |
| Mistral (EU) | $1.54 | $4.84 | $2.37 | $0.154 | 1.05M | NVFP4 | 0.91 s | 132 tok/s | – |
| Amazon Bedrock (US-EAST-1) | $1.68 | $5.28 | $2.58 | $0.312 | 1.05M | Not listed | 1.48 s | 110 tok/s | – |
| Baseten (fp8, Fast) | $2.10 | $6.60 | $3.23 | $0.21 | 1.05M | FP8 | 0.43 s | 176 tok/s | – |
| Baseten (fp8, US, Fast) | $2.10 | $6.60 | $3.23 | $0.21 | 1.05M | FP8 | 0.44 s | 179 tok/s | – |
| Fireworks (Fast), down when checked | $2.10 | $6.60 | $3.23 | $0.39 | 1.05M | Not listed | 1.03 s | 122 tok/s | – |
| Parasail (Fast) | $2.10 | $6.60 | $3.23 | $0.26 | 1.05M | FP8 | 0.86 s | 72 tok/s | – |
| Alibaba Cloud Int. (Prime) | $2.80 | $8.80 | $4.30 | $0.56 | 1M | Not listed | 0.89 s | 108 tok/s | – |

Blended is three parts input price to one part output, per million tokens. Latency is time to first token and throughput is output tokens per second, both averaged from OpenRouter's daily figures for the last 7 days. Cheapest and fastest only count listings that were up when checked.

Wafer is an ernd partner: 10% back on Wafer inference with the ernd card. With ernd card is the blended price less 10% cash back on Wafer charges. It comes back on the card bill and doesn't change Wafer's list price. It applies when you pay Wafer directly, not through OpenRouter credits.

## Questions

### What is the cheapest GLM 5.3 API provider?

Wafer (US), at $0.402 per million tokens blended ($0.0699 input, $1.40 output), in OpenRouter's listings on November 7, 2026.

### How much does GLM 5.3 cost per million tokens?

Input runs from $0.0699 to $2.80 per million tokens across 47 listings, and output from $1.40 to $8.80. The middle listing comes to $1.83 blended.

### Which GLM 5.3 provider is fastest?

Baseten (fp8, US, Fast) had the most throughput over the last 7 days, about 179 tokens per second. Modal was quickest to the first token at 0.34 s.

### How do I pay for GLM 5.3 inference?

Buy OpenRouter credits and it sends each request to one of these providers, or open an account with a provider and pay it directly. The ernd card is a credit card for US companies. It earns 4% back on AI vendor spend and 1.5% back on everything else. Wafer charges on the card earn 10% back.

Source: OpenRouter public model and provider listings (https://openrouter.ai/models), checked November 7, 2026.

The ernd card is a credit card for US companies. It earns 4% back on AI vendor spend and 1.5% back on everything else. [Request the card at app.ernd.ai](https://app.ernd.ai/login?next=%2Fapp%2Fcard%2Fapply&utm_source=llms&utm_campaign=inference_prices).