# DeepSeek Coder V2-0724

DeepSeek Coder V2-0724 is an AI model released by DeepSeek on Jul 24 2024. It has 236B parameters, a 128k context window and open weight. Tracked results include 19% on SWE-Bench Verified and 39.7% on LiveCodeBench.

## Facts

| Field | Value |
| --- | --- |
| Model | DeepSeek Coder V2-0724 |
| Developer | DeepSeek |
| Release date | Wednesday, Jul 24 2024 |
| Licensing | Open Weight |
| Parameters | 236B |
| Context window | 128k |

## Tracked benchmark scores

| Benchmark | Score | Source | What it measures |
| --- | --- | --- | --- |
| SWE-Bench Verified | 19% | Lab | Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. |
| LiveCodeBench | 39.7% | Lab | Coding problems published so recently the AI can't have seen them in training — a contamination-free test of raw programming skill. Higher is better. |
| HumanEval | 87.2% | 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. |

## About DeepSeek Coder V2-0724

DeepSeek Coder V2-0724 was the July 24, 2024 update to DeepSeek's programming API. It continued the Coder V2 line before DeepSeek combined its capabilities with the chat model in V2.5.

DeepSeek subsequently published the exact checkpoint's weights and named it in V2.5's comparison table. The benchmark values here come from that explicitly labeled comparison column, rather than the original June Coder V2 release.

## Questions and answers

### When was DeepSeek Coder V2-0724 released?

DeepSeek Coder V2-0724 was released by DeepSeek on Wednesday, Jul 24 2024.

### Who made DeepSeek Coder V2-0724?

DeepSeek Coder V2-0724 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 Coder V2-0724 get?

DeepSeek Coder V2-0724 reports 3 tracked benchmark scores — SWE-Bench Verified: 19%; LiveCodeBench: 39.7%; HumanEval: 87.2%. Tracked scores may come from lab reports or independent benchmarks; source details accompany the benchmark data.

### What is the context window of DeepSeek Coder V2-0724?

DeepSeek Coder V2-0724 has a context window of 128k. Providers may specify separate input and output token limits.

### How many parameters does DeepSeek Coder V2-0724 have?

DeepSeek Coder V2-0724 is reported at 236B parameters.

### Is DeepSeek Coder V2-0724 open source?

Partly. DeepSeek Coder V2-0724 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 Coder V2-0724?

DeepSeek's previous tracked release was DeepSeek V2-0628 on Jun 28 2024, 26 days earlier. It was followed by DeepSeek V2.5 on Sep 5 2024.


---

Canonical page: https://aireleasetracker.com/model/deepseek/deepseek-coder-v2-0724
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.
