DeepSeek MoE 16B Chat is an open-source chat model created by DeepSeek. It takes text as input and generates text as output, designed for conversational tasks.
Input
Output
Context
4K
Max Output
-
Parameters
16.4B
Input Modalities
Output Modalities
Download 32.8 GB · Runs on vLLM v0.32.0 · 11 GGUF builds
| Precision | Weights | One request at 4K tokens | Smallest GPU that fits |
|---|---|---|---|
| FP16 / BF16 | 30.5 GB | 32.6 GB | 1x NVIDIA A100 SXM 40GB (40 GB) |
| INT8 | 15.3 GB | 16.6 GB | 1x NVIDIA GeForce RTX 3090 (24 GB) |
| INT4 | 7.63 GB | 8.57 GB | 1x NVIDIA T4 (16 GB) |
Computed by Inferbase from the 16,375,728,128 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 10, 2026.
Answered from the facts on this page, with their sources.
DeepSeek MoE 16B Chat has a context window of 4,096 tokens (source: Creator or provider API). A provider may serve less; Inferbase routes on what each provider serves.
Yes. DeepSeek MoE 16B Chat is an open-weight model under the Other license. You can download and self-host it within the license terms.
DeepSeek MoE 16B Chat has 16,375,728,128 parameters (16.4B), as stated in the creator's Hugging Face repository.
Limitations: Short context window: 4,096 tokens. Each is computed from the model's stated facts and benchmark results.
DeepSeek MoE 16B Chat 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 DeepSeek MoE 16B Chat yet, so it is shown as unmeasured, not estimated.
Not on Inferbase Managed today. Add your own provider key or endpoint that serves DeepSeek MoE 16B Chat, and Inferbase runs it on your account with the same routing and logs.
The weights take 30.5 GB at FP16 / BF16, 15.3 GB at INT8, 7.6 GB at INT4, computed from the 16,375,728,128 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.
Start on the free plan with 5,000 requests a month, no card required.