# DeepSeek V2 Lite

DeepSeek V2 Lite is an AI model released by DeepSeek on May 16 2024. It has 16B parameters, a 32k context window and open weight.

## Facts

| Field | Value |
| --- | --- |
| Model | DeepSeek V2 Lite |
| Developer | DeepSeek |
| Release date | Thursday, May 16 2024 |
| Licensing | Open Weight |
| Parameters | 16B |
| Context window | 32k |

## Tracked benchmark scores

| Benchmark | Score | Source | What it measures |
| --- | --- | --- | --- |
| HumanEval | 57.3% | Lab | 164 small Python problems: the AI is given a function's description and has to write the working function. This was the coding benchmark of the GPT-3.5 and GPT-4 era, before the field moved to fixing real bugs in real repositories. Higher is better. |
| MATH | 27.9% | Lab | accuracy on the original competition mathematics test set |
| MMLU | 55.7% | Lab | A 57-subject multiple-choice exam — history, law, medicine, maths — that was the standard measure of how much a model knows from 2020 until roughly 2024, when frontier scores crowded into the high 80s and labs moved on to harder tests. The scores here were published years apart under different testing setups, so read them as a historical record rather than a like-for-like ranking. Higher is better. |
| GSM8K | 72% | Lab | Grade-school maths word problems that take a few steps of arithmetic to work through. It separated the models of 2022 and 2023 sharply, then saturated. One caveat on the historical numbers: OpenAI included part of the GSM8K training set in GPT-4's pre-training mix, so GPT-4's score is not a clean few-shot result. Higher is better. |

## About DeepSeek V2 Lite

DeepSeek V2 Lite followed the original V2 release on May 16, 2024. It brought Multi-head Latent Attention and sparse expert routing to a smaller model intended to make the architecture easier to study and deploy.

The release included base and chat versions with a 32K context window. The scores shown here use the chat checkpoint; they should not be read as results for the larger 236B V2 model.

## Questions and answers

### When was DeepSeek V2 Lite released?

DeepSeek V2 Lite was released by DeepSeek on Thursday, May 16 2024.

### Who made DeepSeek V2 Lite?

DeepSeek V2 Lite was built by DeepSeek. Chinese AI lab known for efficient, open-weight models. Gained attention for strong performance at lower cost.

### What benchmark scores did DeepSeek V2 Lite get?

DeepSeek V2 Lite reports 4 tracked benchmark scores — HumanEval: 57.3%; MATH: 27.9%; MMLU: 55.7%; GSM8K: 72%. Tracked scores may come from lab reports or independent benchmarks; source details accompany the benchmark data.

### What is the context window of DeepSeek V2 Lite?

DeepSeek V2 Lite has a context window of 32k. Providers may specify separate input and output token limits.

### How many parameters does DeepSeek V2 Lite have?

DeepSeek V2 Lite is reported at 16B parameters.

### Is DeepSeek V2 Lite open source?

Partly. DeepSeek V2 Lite is an open-weight model: the trained weights are free to download and run locally or on your own infrastructure, but the training data and code are not fully released and the license may restrict some commercial uses. It is not open source in the strict sense.

### What came before and after DeepSeek V2 Lite?

DeepSeek's previous tracked release was DeepSeek V2 on May 6 2024, 10 days earlier. It was followed by DeepSeek Coder V2 on Jun 14 2024.


---

Canonical page: https://aireleasetracker.com/model/deepseek/deepseek-v2-lite
Site index: https://aireleasetracker.com/llms.txt
Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.
