GreenPT / greenpt/glm-5.3-flash-eu
GLM-5.3 Flash - access through LLMTR
GLM-5.3 Flash is the cheapest chat row on this provider and still offers a 1,000,000-token context window. It genuinely processes image input: in measurement it described a two-photo product image correctly and its input token count rose by the image's own tokens. Tool/function calling was verified over a full two-turn loop. Repeated prompt prefixes are served from a cache and those input tokens bill at a fifth of the normal input rate. It costs ten times less per token than the larger GLM-5.3 in the same family, which makes it a fit for high-volume classification, summarisation and image tagging. The model reasons by default and those tokens bill as output. Schema-enforced JSON output via `response_format` was rejected by this row in measurement; use Kimi K2.7 Code on this provider when you need schema enforcement.
Technical specifications
| Canonical ID | greenpt/glm-5.3-flash-eu |
|---|---|
| Provider | GreenPT |
| Context window | 1,000,000 tokens |
| Operations | CHAT_COMPLETIONS |
| Modalities | text, image |
Pricing
An 8% platform margin applies to credit top-ups; model usage prices are not separately marked up.
| Operation | Metric | Unit | Price |
|---|---|---|---|
| CHAT_COMPLETIONS | INPUT_TEXT | PER_1M_TOKENS | $0.134234 |
| CHAT_COMPLETIONS | CACHE_READ | PER_1M_TOKENS | $0.026847 |
| CHAT_COMPLETIONS | OUTPUT_TEXT | PER_1M_TOKENS | $0.536936 |
Example usage
With existing OpenAI SDK flows, change only the base URL and model identifier.
curl https://llmtr.com/v1/chat/completions -H "Authorization: Bearer llmtr-your_key" -H "Content-Type: application/json" -d '{"model":"greenpt/glm-5.3-flash-eu","messages":[{"role":"user","content":"Hello"}]}'Guides about this model
- How to Choose a GDPR-Compliant LLM API: A Developer Checklist - Approaches LLM API selection for a GDPR-scoped application through data processing, the sub-processor list and what can actually be verified.
- How to Verify an EU Data Residency Claim for an AI API - Breaks an AI API's European hosting claim into evidence types and shows which part an engineering team can actually verify.
- No Training on Inputs: How to Read That Commitment - Reads the no-training-on-inputs commitment by separating it from neighbouring clauses such as retention period and abuse monitoring.
- EU-Hosted LLM Models: Reading a 38-Row Catalog - Separates the EU category's 38 rows into chat, embedding and reranking, and gives a selection framework based on measured capabilities.
- How to Use GreenPT Models Through LLMTR - Covers using GreenPT rows through LLMTR, including the hidden system instruction and default reasoning that drive cost.
- Choosing an LLM API as a Small EU Company or Startup - Lists the binding criteria, the deferrable decisions and the first integration steps for small EU teams choosing an LLM API.
Related models
- Kimi K3 greenpt/kimi-k3-eu
- GLM-5.3 greenpt/glm-5.3-eu
- Kimi K2.7 Code greenpt/kimi-k2.7-code-eu
- Kimi K2.6 greenpt/kimi-k2.6-eu
- MiniMax M2.5 greenpt/minimax-m2.5-eu
- DeepSeek V4 Flash greenpt/deepseek-v4-flash-0731
- GLM-5.2 greenpt/glm-5.2-eu
- Devstral 2 123B greenpt/devstral-2-123b-instruct-2512