Qwen 2.5 3B Instruct is an open-source, text-in, text-out chat model created by Qwen.
Input
Output
Context
33K
Max Output
8K
Parameters
3.1B
Input Modalities
Output Modalities
Some values come from web research and may be incorrect. Report incorrect data
None the catalog can show from stated facts or measured scores.
Download 2.69 GB · Runs on vLLM v0.32.0
| Precision | Weights | One request at 33K tokens | Smallest GPU that fits |
|---|---|---|---|
| FP16 / BF16 | 5.75 GB | 9.07 GB | 1x NVIDIA T4 (16 GB) |
| INT8 | 2.87 GB | 6.05 GB | 1x NVIDIA T4 (16 GB) |
| INT4 | 1.44 GB | 4.54 GB | 1x NVIDIA T4 (16 GB) |
Computed by Inferbase from the 3,085,938,688 parameters in the creator's Hugging Face repository; one request adds its KV cache, activations and framework overhead, with the layer layout estimated from the parameter count.
Public resources
Specification sources
As of Nov 13, 2026.
Answered from the facts on this page, with their sources.
Qwen 2.5 3B Instruct has a context window of 32,768 tokens and a maximum output of 8,192 tokens (source: Creator or provider API). A provider may serve less; Inferbase routes on what each provider serves.
Yes. Qwen 2.5 3B Instruct is an open-weight model under the Other license. You can download and self-host it within the license terms.
Qwen 2.5 3B Instruct has 3,085,938,688 parameters (3.1B), as stated in the creator's Hugging Face repository.
Qwen 2.5 3B Instruct accepts Text and produces Text. The Capabilities card lists the API features it supports, such as function calling and structured output.
No benchmark source has published a result for Qwen 2.5 3B Instruct yet, so it is shown as unmeasured, not estimated.
Not on Inferbase Managed today. Add your own provider key or endpoint that serves Qwen 2.5 3B Instruct, and Inferbase runs it on your account with the same routing and logs.
The weights take 5.7 GB at FP16 / BF16, 2.9 GB at INT8, 1.4 GB at INT4, computed from the 3,085,938,688 parameters its creator publishes. Each request adds its KV cache and the framework's overhead. The hardware section shows one request and the smallest GPU that fits; the GPU planner sizes it for your context length and traffic.
The model card of Qwen 2.5 3B Instruct lists one language: English. That is what Qwen states; the catalog does not test language quality.
Start on the free plan with 5,000 requests a month, no card required.