GLM 4.5 Air:thinking API
Zhipu's GLM 4.5 Air:thinking costs $0.13 per million input tokens and $0.851 per million output tokens from 1 supplier. Each request through the OpenAI-compatible API goes to the offer with the lowest estimated charge that meets your rules.
- Providers compete for each request. The exchange ranks the offers that meet your rules by estimated charge. It fixes your maximum charge before the first token, and you pay for the tokens that the model uses.
- The OpenAI and Anthropic APIs. Use the OpenAI or Anthropic SDK that you have. Set the base URL and your key, and send glm-4.5-air:thinking as the model.
- Your data rules. Send x-liquid-retention: none and only offers whose provider does not keep your content compete for the request.
About GLM 4.5 Air:thinking
GLM 4.5 Air:thinking is a language model from Zhipu. Its release date is 28 July 2025. Lighter GLM-4.5 variant for fast coding assistance and cheaper agents. Its weights are open. Its training data ends in April 2025. It takes text and gives text. The largest offer takes 128,000 input tokens. It has 1 supplier on this exchange now.
Description, release date and licence: models.dev (https://models.dev/)
Facts
- Model name
- glm-4.5-air:thinking
- Maker
- Zhipu
- Context window
- 128K tokens
- Input price
- $0.13 per million tokens
- Output price
- $0.851 per million tokens
- Suppliers
- 1 supplier
- Regions
- us
Providers
- Architect: us, declared precision provider-managed, $0.13 input and $0.851 output per million tokens, $0.025 cache read per million tokens, 128K tokens of context
Scores
- HLE: 7.0, #292 of 465
- GPQA: 73.3, #243 of 432
- AA-LCR: 46.7, #274 of 409
- Terminal-Bench Hard: 20.5, #162 of 311
- LiveCodeBench: 68.4, #57 of 250
- AIME 2025: 80.7, #35 of 199
- MMLU-Pro: 81.5, #83 of 253
- τ²-Bench: 46.5, #185 of 313
- IFBench: 37.6, #243 of 321
Scores: Artificial Analysis (https://artificialanalysis.ai/)
Request
curl https://router.inference.ai.exchange/v1/chat/completions \
-H "Authorization: Bearer $LIQUID_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "glm-4.5-air:thinking",
"messages": [{"role":"user","content":"Say hello."}]
}'Questions
- What is GLM 4.5 Air:thinking?
- GLM 4.5 Air:thinking is a language model from Zhipu. Its release date is 28 July 2025. Lighter GLM-4.5 variant for fast coding assistance and cheaper agents. Its weights are open. Its training data ends in April 2025.
- How much does GLM 4.5 Air:thinking cost?
- Offers now ask $0.13 per million input tokens and $0.851 per million output tokens. A job pays for the tokens it uses. The exchange picks the cheapest eligible offer.
- What is the context length of GLM 4.5 Air:thinking?
- The largest offer takes 128,000 input tokens. A smaller offer can take less. The exchange skips an offer that cannot take your message.
- What can GLM 4.5 Air:thinking do?
- Its providers list Reasoning effort, JSON object, JSON schema, Frequency penalty, Logit bias, Log probabilities, Min-p sampling, Presence penalty, Repetition penalty, Seed, Stop sequences, Top-k sampling, Top-p sampling, Text, Tools.
- What inputs and outputs does GLM 4.5 Air:thinking support?
- Text → Text.
- Which providers serve GLM 4.5 Air:thinking?
- It has 1 supplier now, with 1 offer. The table above ranks them by the cost of one agent-mix job.
- How do I call GLM 4.5 Air:thinking?
- Send "glm-4.5-air:thinking" as the model in an OpenAI or Anthropic request. The API section links to setup steps for each client.
- When was GLM 4.5 Air:thinking released?
- Its creator released it on 28 July 2025.
- When was GLM 4.5 Air:thinking listed?
- The exchange first listed it on 10/10/2026. This is not its release date.
Related pages
- GLM 5.3 Flash
- GLM 5.3
- Autoglm: Phone 9B Multilingual
- GLM 5.2
- GLM 5.3 Flash Uncensored
- GLM 4.7 Flash:thinking
- Popular open weights
- Reasoning models
- Set up a client
- All models
- Providers
- Discover the best model for your budget
- Quickstart
- llms.txt: the index of every page, as Markdown
- llms-full.txt: every page in one file
- This page as Markdown