LLMTR / llmtr/qwen3-5-4b
Qwen 3.5 4B - access through LLMTR
LLMTR Qwen 3.5 4B is a first-party language model served from Turkey. It is invoked through the OpenAI-compatible Chat Completions API and requests are never forwarded to a third-party provider. It offers a 128K context window (prompt and completion together, per request). This is a small, fast model: a good fit for Turkish chat, summarization, classification, RAG chatbots, and agent flows that need tool calling. It accepts text input only; image, audio and video parts are refused with a 400. Repeating a prompt prefix hits the prompt cache, and cached tokens are billed at a discounted rate, which meaningfully lowers cost for flows with long system prompts. Thinking is on by default: the model produces a reasoning pass before answering, returns it to you in the `reasoning_content` field, and those tokens are spent from your `max_tokens` budget. Set `max_tokens` to at least 2048 on a default call — with a smaller budget the reasoning consumes it and the content comes back empty. To turn thinking off use `llmtr/qwen3-5-4b:fast` or send `reasoning: false` in the request body.
Technical specifications
| Canonical ID | llmtr/qwen3-5-4b |
|---|---|
| Provider | LLMTR |
| Context window | 131,072 tokens |
| Operations | CHAT_COMPLETIONS |
| Modalities | text |
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 | $1.00 |
| CHAT_COMPLETIONS | CACHE_READ | PER_1M_TOKENS | $0.250000 |
| CHAT_COMPLETIONS | OUTPUT_TEXT | PER_1M_TOKENS | $3.00 |
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":"llmtr/qwen3-5-4b","messages":[{"role":"user","content":"Hello"}]}'Guides about this model
- Combining reasoning toggle and tool calling on LLMTR Qwen 3.5 4B - LLMTR Qwen 3.5 4B carries both tool calling and a thinking mode that can be toggled on or off. This article covers how to use these two capabilities together in a task.
- Trae Agent with LLMTR: Python configuration and trajectory records - Trae Agent reaches LLMTR through a dispatching class that splits models across two client implementations. This guide walks the Python configuration, the routing decision and the fields visible in the run record.
- The models.dev LLMTR provider entry and metadata sync - models.dev is not a service but an open metadata database holding model ids and capabilities in TOML files. We compare the current LLMTR entry against the live catalog to see which rows are missing and which one is retired.
Related models
- Gemma 4 llmtr/gemma-4
- Qwen 3.6 35B-A3B llmtr/qwen3-6-35b
- Trendyol Asure 12B llmtr/trendyol-asure-12b
- Magibu 11B v8 llmtr/magibu-11b-v8
- Muse Glimmer 30B (Turkey) llmtr/muse-glimmer-30b-tr
- EmbeddingGemma 300M llmtr/embeddinggemma-300m
- GPT-6 Astra openai/gpt-6-astra
- GPT-6 Sol openai/gpt-6-sol