whatmodel.sh / DeepSeek-Coder-V2 Lite Instruct (16B-A2.4B)

DeepSeek-Coder-V2 Lite Instruct (16B-A2.4B)

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

DeepSeek-Coder-V2 Lite Instruct (16B-A2.4B) is an open-weights language model from DeepSeek, released June 2024. Its published size is 15.71B parameters (2.4B active per token) and a 128k-token context window. Licence: deepseek.

Catalogue identity notes

Maker repo deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct @e434a23: licence tag other (deepseek-license). 15,706,484,224 parameters from the Hugging Face safetensors metadata. Mixture-of-experts: pinned README: 'DeepSeek-Coder-V2 with 16B and 236B parameters ... which has actived parameters of only 2.4B and 21B'. Released 2024-06: month the maker's Hugging Face repo was created (createdAt 2024-06-14). Ollama tags deepseek-coder-v2:16b-lite-instruct-q4_K_M, deepseek-coder-v2:16b-lite-instruct-q8_0 are the library's instruct/chat builds of this checkpoint (tag names on the pinned tags page); download sizes are estimated from the parameter count.

45.0±1.2#443 of 719 LLMsfrom 10 benchmark results across 7 sources

$ how it compares

The LLMs ranked either side of DeepSeek-Coder-V2 Lite Instruct (16B-A2.4B), out of 828 with a score, then more from DeepSeek.

#modelscorewhereRAMreleased
527Phi-3 Medium 128K Instruct Microsoft45±1local~9.8GB2024-04
528QwQ-32B-Preview Alibaba45±2local~21GB2024-11
529DeepSeek-Coder-V2 Lite Instruct (16B-A2.4B) DeepSeek45±1local~11GB2024-06
530GPT-4o Audio Preview (2024-10-01) OpenAI45±3comparison only2024-10
531Claude 2.1 Anthropic45±1comparison only2023-11
more from DeepSeek
21DeepSeek V4.1 Flash DeepSeek78±1api $0.26/M +local~537GB2026-09
31DeepSeek V4 Pro (0813) DeepSeek76±1api $1.63/M +local~1010GB2026-08
35DeepSeek V3.2 Speciale DeepSeek76±3comparison only

see all LLMs · find the ones that fit your machine

$ what it's good at

agent tasks▃▁▃▆█▆▅▅▅▄▃▁ #297 of 300 * · 2.8% of its score · 1 test

Beats QwQ-32B-Preview here, though they rank higher overall. Trails Granite 4.0 350M, Qwen2.5 0.5B Instruct, Granite 4.0 H 350M here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
BFCL v321.0% ✎▁▃▁▁▁▃▅█▅▄▁▁ beats 4/114medium · avg 52%▇ 2.8%blunt · at its level
coding▁▁▁▃▆██▇▆▄▃▁ #312 of 458 · 34% of its score · 6 tests

Beats Solar Open 100B (102B-A12B), GLM-4.7-Flash, Qwen3 4B Instruct 2507 here, though they rank higher overall. Trails DeepSeek LLM 67B Chat, WizardLM 70B V1.0, Phi-4-mini 3.8B here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
LiveCodeBench official (Jan 1–Nov 1, 2024)20.6% ✎▁▁·▁▂▄▆█▅▂·▂ beats 5/31hard · avg 26%▇▇▇▇▇▇▇▇ 20%sharp
WebApp1K-React pass@1 (onekq-ai)46.1% ✎▃▁▁█▁▄▅▃▄▇▁▄ beats 11/31medium · avg 44%▇▇ 4.7%sharp · at its level · noisy
CodeReviewQA: solution identification (hard), invariant accuracy12.8% ✎▃··▁▂▃▃▂█▃▄▂ beats 10/47medium · avg 54%▇ 3.4%sharp · at its level · noisy
CodeReviewQA: change localisation (hard), invariant accuracy6.2% ✎█▂▁▁▂▃▂▃▅▅▅▇ beats 15/47medium · avg 36%▇ 3.1%sharp · noisy
CodeReviewQA: automated code refinement, exact match rate27.0% ✎▁▁▁▁▁▁▂▂▃█▅▄ beats 18/47hard · avg 33%▇ 2.4%at its level
CodeReviewQA: change type recognition, invariant accuracy75.7% ✎▁▁▁······▁▃█ beats 31/47easy · avg 67%▇ 0.1%blunt · at its level · noisy
other tests3 not in a task · 63% of its score
testresultvs the others in its rundifficultycountswhy
LiveBench 2024-07-2629.5% ✎▁▁▁▂▂▃▂█▆▃▅▃ beats 26/61medium · avg 41%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 48%at its level
EvalPerf correctness (118 tasks, 20240328)79.0% ✎▂·▂▂▂▆▅▆▆▆▃█ beats 16/29easy · avg 76%▇▇▇▇ 10%at its level
DSEffi-Bench Pass (1,000 data-science tasks, greedy)17.1% ✎▅··▃▅·█·▃██▃ beats 0/15hard · avg 24%▇▇ 5.5%typical

$ where to run it

llama.cpp · recommended

version

  1. 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 | sh
    Windows
    irm https://llama.app/install.ps1 | iex

    other ways to install ↗

  2. Run DeepSeek-Coder-V2 Lite Instruct (16B-A2.4B) · needs ~10.4 GB · from huggingface.co/lmstudio-community
    chat in the terminal
    llama cli -hf lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q4_K_M

LM Studio

version

  1. Install LM Studio · Open the app once before using the lms command.

    download it (macOS / Windows / Linux) ↗

  2. Run DeepSeek-Coder-V2 Lite Instruct (16B-A2.4B) · needs ~10.4 GB · from huggingface.co/lmstudio-community

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/lmstudio-community/DeepSeek-Coder-V2-Lite-Instruct-GGUF@Q4_K_M -y
  4. Then run lms ls to see its name, and lms chat followed by that name

Ollama

version

  1. Install Ollama

    download it (macOS / Windows) ↗

    Linux
    curl -fsSL https://ollama.com/install.sh | sh
  2. Run DeepSeek-Coder-V2 Lite Instruct (16B-A2.4B) · needs ~9.93 GB · from ollama.com
    chat in the terminal
    ollama run deepseek-coder-v2:16b-lite-instruct-q4_K_M

MLX (Mac)

version

  1. Install MLX (Mac) · Apple silicon Macs only, text only.
    with uv
    uv tool install mlx-lm
    or
    pip install mlx-lm
  2. Run DeepSeek-Coder-V2 Lite Instruct (16B-A2.4B) · needs ~8.8 GB, Mac only · from huggingface.co/mlx-community
    chat in the terminal
    mlx_lm.chat --model mlx-community/DeepSeek-Coder-V2-Lite-Instruct-4bit-mlx
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model mlx-community/DeepSeek-Coder-V2-Lite-Instruct-4bit-mlx

Weights and documentation: huggingface.co/deepseek-ai ↗

$ how the score was worked out · LiveBench model judgments (HF… 48%, LiveCodeBench official websit… 20%, EvalPlus / EvalPerf published… 10% and 4 more

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

sourceresultsshare of the scorelicence
LiveBench model judgments (HF livebench/model_judgment, 2024-10-22 revision) with the LiveBench question datasets1▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 48%Apache-2.0 (stated in the LiveBench datasheet)
LiveCodeBench official website results (2024-12-26)1▇▇▇▇▇▇ 20%CC-BY-SA-4.0
EvalPlus / EvalPerf published evaluation runs1▇▇▇ 10%Apache-2.0
CodeReviewQA: The Code Review Comprehension Assessment for Large Language Models (arXiv:2503.16167v2)4▇▇▇ 9%CC-BY-4.0
DSEffi-Bench: Demystifying Large Language Models' Capability in Efficient Data Science Code Generation — authors' own model runs1▇▇ 5%CC BY 4.0
WebApp1K models leaderboard (onekq-ai)1▇ 5%MIT
Berkeley Function Calling Leaderboard result archive1▇ 3%Apache-2.0
prior · pull toward the middle▇ 1%
weighted average= 45.0

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

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