whatmodel.sh / DeepSeek-Coder-V2 Instruct (0724)
DeepSeek-Coder-V2 Instruct (0724)
↗ share
+ contribute
DeepSeek-Coder-V2 Instruct (0724) is an open-weights language model from DeepSeek, released September 2024. Its published size is 235.74B parameters (21B active per token). Licence: other (deepseek-license).
Catalogue identity notes
Maker repo deepseek-ai/DeepSeek-Coder-V2-Instruct-0724 @80378f7: licence tag other (deepseek-license). 235,741,434,880 parameters from the Hugging Face safetensors metadata; active 21B per token (DeepSeek-Coder-V2 card: 236B total, 'actived parameters of only ... 21B'). Released 2024-09: month the maker's Hugging Face repo was created (createdAt 2024-09-05). DeepSeek API updates page (api-docs.deepseek.com/updates): 'Date: 2024-07-24 deepseek-coder: The deepseek-coder model has been upgraded to DeepSeek-Coder-V2-0724' (superseded by DeepSeek-V2.5 on 2024-09-05). Distinct from the catalogued deepseek-coder-v2 (0614).
$ how it compares
The LLMs ranked either side of DeepSeek-Coder-V2 Instruct (0724), out of 828 with a score, then more from DeepSeek.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 358 | DeepSeek-Coder-V2 Instruct (0614) | 53±1 | local | ||
| 359 |  s1.1 32B | 53±1 | local | ||
| 360 | DeepSeek-Coder-V2 Instruct (0724) | 53±1 | local | ||
| 361 |  Olmo 3 7B Think | 53±1 | local | ||
| 362 |  GPT-4.1 nano | 53±1 | api $0.17/M | ||
| more from DeepSeek | |||||
| 21 | DeepSeek V4.1 Flash | 78±1 | api $0.26/M +local | ||
| 31 | DeepSeek V4 Pro (0813) | 76±1 | api $1.63/M +local | ||
| 35 | DeepSeek V3.2 Speciale | 76±3 | comparison only | ||
$ what it's good at
agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #233 of 300 * · 2.0% of its score · 1 test
Beats o3-mini, Ministral 3 14B Reasoning 2512, DeepSeek R1 0528 Qwen3 8B here, though they rank higher overall. Trails Granite 4.0 1B, SmolLM3 3B, Phi-3.5-mini-instruct here, though they rank lower overall.
coding▁▁▁▃▆██▇▆▄▃▁ #191 of 458 * · 38% of its score · 2 tests
Beats Gemini 2.5 Flash, Qwen3.5 4B, Qwen3 30B-A3B Instruct 2507 here, though they rank higher overall. Trails QwQ-32B-Preview here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| CRUXEval (ZeroEval, 800 problems) | 69.5% ✎ | ▁▂▆▃▆█▆▃▂·▁▁ beats 27/47 | 22% |
| Aider code editing (133 Exercism Python exercises, two attempts) | 72.9% ✎ | ▁·▁▂▄▁▆▄█▆▂▂ beats 55/62 | 16% |
math▁▂▃▅▆▅▄█▄▁▁▁ #280 of 398 * · 7.6% of its score · 1 test
Beats Molmo 2 8B here, though they rank higher overall. Trails DeepSeek R1 Distill Qwen 1.5B, Granite 8B Code Base (4K), Mixtral 8x7B v0.1 (base) here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| GSM8K (ZeroEval, zero-shot) | 91.5% ✎ | ▁▄·▁▄█▅▅▅███ beats 24/43 | 7.6% |
other tests2 not in a task · 52% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| MMLU-Redux (ZeroEval, 2,778 questions) | 80.2% ✎ | ▁·▁▁▄▄▄▅█▄▂▁ beats 28/53 | 34% |
| ZebraLogic (ZeroEval, 1,000 puzzles) | 20.5% ✎ | ▁▁▃▆██▁▁·▁▁▁ beats 31/57 | 17% |
$ where to run it
llama.cpp · recommended
version
- Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.macOS / Linux
curl -LsSf https://llama.app/install.sh | shWindowsirm https://llama.app/install.ps1 | iex - Run DeepSeek-Coder-V2 Instruct (0724) · needs ~142.5 GB · from huggingface.co/lmstudio-communitychat in the terminal
llama cli -hf lmstudio-community/DeepSeek-Coder-V2-Instruct-0724-GGUF:Q4_K_M - Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf lmstudio-community/DeepSeek-Coder-V2-Instruct-0724-GGUF:Q4_K_M
LM Studio
version
- Install LM Studio · Open the app once before using the lms command.
- Run DeepSeek-Coder-V2 Instruct (0724) · needs ~142.5 GB · from huggingface.co/lmstudio-community
- Or download it in the terminal
lms get https://huggingface.co/lmstudio-community/DeepSeek-Coder-V2-Instruct-0724-GGUF@Q4_K_M -y - Then run lms ls to see its name, and lms chat followed by that name
Other ways to run it
- Run it as published · needs ~471.49 GB · from huggingface.co/deepseek-ai
uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='deepseek-ai/DeepSeek-Coder-V2-Instruct-0724', torch_dtype='auto', device_map='auto', trust_remote_code=True)('Once upon a time', max_new_tokens=40)[0]['generated_text'])"
Weights and documentation: huggingface.co/deepseek-ai ↗
$ how the score was worked out · WildEval ZeroEval published r… 81%, Aider code editing benchmark… 16%, Berkeley Function Calling Lea… 2%
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
 WildEval ZeroEval published results | 4 | 81% |
 Aider code editing benchmark runs (old edit leaderboard) | 1 | 16% |
| Berkeley Function Calling Leaderboard result archive | 1 | 2% |
| prior · pull toward the middle | 0% | |
| weighted average | = 53.0 |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json