OpenRouter
Cheaper right now$0.075 / $0.25 per million input / output tokens
The model name is z-ai/glm-5.3-flash, using OpenRouter's standard API URL.
GLM 5.3 Flash: lower AI costs
At current list prices, GLM 5.3 Flash on OpenRouter costs clearly less than the other three recommended models for the same input and output tokens.
Prices checked on 2026-09-19. Providers may change prices at any time; the billing page is always authoritative.
Price per million tokens
To make the comparison easier, Qwen uses list prices for the China North 2 (Beijing) region, converted to USD at $1 ≈ ¥7.26. Your actual bill is charged in the platform's own currency.
| Model | Input | Output | Highlights | vs. GLM |
|---|---|---|---|---|
z-ai/glm-5.3-flashLowest now
OpenRouter, currently 50% off
|
$0.075 | $0.25 | Great value, slightly stronger reasoning | 1× |
qwen3.8-flash
Alibaba Cloud Model Studio, China North 2 (Beijing)
|
¥0.80, about $0.110 | ¥2.70, about $0.372 | Great value, faster responses | Input about 1.5×, output about 1.5× |
gpt-5.6-luna
OpenAI official
|
$0.20 | $1.20 | Best quality, most reliable | Input about 2.7×, output about 4.8× |
qwen3.7-plus
Alibaba Cloud Model Studio, input up to 256K
|
¥2.00, about $0.276 | ¥8.00, about $1.101 | Best quality, most reliable | Input about 3.7×, output about 4.4× |
What do 10,000 auto-replies cost?
The estimates below use exactly the same token counts and compare text costs only. Web search, images and other tool calls are not included.
GLM 5.3 FlashOpenRouter current discounted price (50% off)
$2.25
Qwen3.8 FlashAlibaba Cloud Model Studio, China North 2 (Beijing) list price
¥24.10about $3.32
GPT-5.6 LunaOpenAI official standard price
$7.60
Qwen3.7 PlusAlibaba Cloud Model Studio, China North 2 (Beijing) list price
¥64.00about $8.82
GLM is also available through a China-based API
Alibaba Cloud Model Studio offers GLM 5.3 Flash supplied directly by Zhipu. It is easier to reach from mainland China and bills in RMB, but it currently costs more than OpenRouter's limited-time discount.
Alibaba Cloud Model Studio
Billed in RMB¥0.8 / ¥2.8 per million input / output tokens
The model name is ZHIPU/GLM-5.3-Flash. The API URL must include your own WorkspaceId.
Why your actual cost may be higher:
GLM 5.3 Flash uses thinking tokens, which are billed at the output price. Conversation history, long business documents, retries and image inputs also use more tokens.
Official pricing sources
These fees are not charged by TGEgg. All API usage is billed to your own model provider account.
Picked a model? Start configuring
Have an API key from that provider ready. TGEgg runs a text and image test before saving.