- Size
- 15.7B parameterscounted from the safetensors header on the Hub
- Licence
- custom, read it, DeepSeek License AgreementOur reading, check the licence: A custom licence we have not read for this model: other makers use the same name for different texts. Read the model's own licence file before you use it.Licence text
- On the Hub since
- 14 Jun 2024the date the repo was created
- Context length
- 160k tokensfrom the model's config
- Kind
- Base model (no base model declared on the Hub)DeepseekV2ForCausalLM
- On Hugging Face
- deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct86k downloads in the last 30 days, 718 likes
Scores
- default setting87.6%Qwen2.5-Coder Technical Report
Epoch's name: DeepSeek-Coder-V2-Lite-Instruct. Best closed model on this set: gpt-4-0314, 92.0% (OpenAI). All 170 results
Claimed by the maker Claimed
The model card has no claimed scores in a machine-readable form (the card's model-index), so none are shown. The maker may give scores in the card text: read the card.
Other views, which we link to and do not copy: look this model up on Artificial Analysis and the Arena leaderboard.
Independent scores from Epoch AI (CC-BY 4.0), fetched 19 Nov 2026: Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from ‘https://epoch.ai/benchmarks’ [online resource]. Each row links its own source. What the badges mean.
Run it yourself
Memory needed = weights + context cache + 0.50 GB for the runtime. Context cache at 8k tokens: 0.24 GB, worked out from the model's config (deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct), at 16-bit precision. Choose a machine above to see a verdict for each format. Method
GGUF for llama.cpp, Ollama, LM Studio and others
lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF (converted by lmstudio-community), 14k downloads. Sizes are the real files on the Hub, all parts added up.
| Format | Size | Memory needed | Get it |
|---|---|---|---|
| IQ3_M | 7.0 GBfile size | 7.8 GB7.0 + 0.24 + 0.50 | Commandsllama.cpp llama-server -hf lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF:IQ3_MOllama ollama run hf.co/lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF:IQ3_MDownload hf download lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF --include "DeepSeek-Coder-V2-Lite-Instruct-IQ3_M.gguf" --local-dir DeepSeek-Coder-V2-Lite-Instruct-GGUF |
| Q3_K_L | 7.9 GBfile size | 8.6 GB7.9 + 0.24 + 0.50 | Commandsllama.cpp llama-server -hf lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q3_K_LOllama ollama run hf.co/lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q3_K_LDownload hf download lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF --include "DeepSeek-Coder-V2-Lite-Instruct-Q3_K_L.gguf" --local-dir DeepSeek-Coder-V2-Lite-Instruct-GGUF |
| IQ4_XS | 8.0 GBfile size | 8.7 GB8.0 + 0.24 + 0.50 | Commandsllama.cpp llama-server -hf lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF:IQ4_XSOllama ollama run hf.co/lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF:IQ4_XSDownload hf download lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF --include "DeepSeek-Coder-V2-Lite-Instruct-IQ4_XS.gguf" --local-dir DeepSeek-Coder-V2-Lite-Instruct-GGUF |
| Q4_K_M | 9.7 GBfile size | 10.4 GB9.7 + 0.24 + 0.50 | Commandsllama.cpp llama-server -hf lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q4_K_MOllama ollama run hf.co/lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q4_K_MDownload hf download lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF --include "DeepSeek-Coder-V2-Lite-Instruct-Q4_K_M.gguf" --local-dir DeepSeek-Coder-V2-Lite-Instruct-GGUF |
| Q5_K_M | 11.0 GBfile size | 11.8 GB11.0 + 0.24 + 0.50 | Commandsllama.cpp llama-server -hf lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q5_K_MOllama ollama run hf.co/lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q5_K_MDownload hf download lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF --include "DeepSeek-Coder-V2-Lite-Instruct-Q5_K_M.gguf" --local-dir DeepSeek-Coder-V2-Lite-Instruct-GGUF |
| Q6_K | 13.1 GBfile size | 13.8 GB13.1 + 0.24 + 0.50 | Commandsllama.cpp llama-server -hf lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q6_KOllama ollama run hf.co/lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q6_KDownload hf download lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF --include "DeepSeek-Coder-V2-Lite-Instruct-Q6_K.gguf" --local-dir DeepSeek-Coder-V2-Lite-Instruct-GGUF |
| Q8_0 | 15.6 GBfile size | 16.3 GB15.6 + 0.24 + 0.50 | Commandsllama.cpp llama-server -hf lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q8_0Ollama ollama run hf.co/lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q8_0Download hf download lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF --include "DeepSeek-Coder-V2-Lite-Instruct-Q8_0.gguf" --local-dir DeepSeek-Coder-V2-Lite-Instruct-GGUF |
MLX for Apple silicon
| Format | Size | Memory needed | Repo |
|---|---|---|---|
| MLX 2bitmasahiroid/DeepSeek-Coder-V2-Lite-Instruct-Pyrite-2bit-MLX | 4.9 GBfile size | 5.6 GB4.9 + 0.24 + 0.50 | Download page |
| MLX 4bitmlx-community/DeepSeek-Coder-V2-Lite-Instruct-4bit | 8.2 GBfile size | 9.0 GB8.2 + 0.24 + 0.50 | Download page |
Original weights size from the safetensors header, for vLLM, SGLang, Transformers
| Format | Size | Memory needed | Repo |
|---|---|---|---|
| BF16original weights | 29.3 GBfile size | 30.0 GB29.3 + 0.24 + 0.50 |
hf download deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct --local-dir DeepSeek-Coder-V2-Lite-InstructOther conversions (4; sizes not read)
- nm-testing/DeepSeek-Coder-V2-Lite-Instruct-FP8 FP8, 654 downloads
- nm-testing/DeepSeek-Coder-V2-Lite-Instruct-quantized.w8a8 Other, w8a8, 48 downloads
- brunopio/DeepSeek-Coder-V2-Lite-Instruct-nbits4-GS0-Axis0-HQQ-T Other, 38 downloads
- huggingkot/DeepSeek-Coder-V2-Lite-Instruct-q4f16_1-MLC Other, 9 downloads
The Hub tags 71 quantised repos as conversions of this model; we read the most downloaded and show the maker's own first.
Run it by API
No provider on Hugging Face Inference Providers serves this model (as of 19 Nov 2026, 13:58 UTC). Other hosts may; we only list what the router reports.
Not built yet: one key for every open model, with BlockRun, and our own tested-quality score per provider. Until then, precision and measured speed are the best guide to a provider here.
Planned for this page Not built yet
- Try it in private chat
- A link to Solana AI's confidential chat for the models it serves. This model is not listed as served there.
- Fine-tune it
- Fine-tuning a few base models from this page and publishing the result here.
- Creators and earnings
- Pages for people who publish fine-tunes and adapters of this model, and what they earn.
- Fund and bounties
- Bounties for new open models, paid from a public fund. How the fund will work.
- Contribute compute
- How to give compute to open model work, with Akash.
Model data from the Hugging Face Hub 19 Nov 2026, 13:43 UTC. The repo. Found a mistake on this page? Tell us.