whatmodel.sh / DeepSeek-Coder 6.7B Base

DeepSeek-Coder 6.7B Base

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

DeepSeek-Coder 6.7B Base is an open-weights language model from DeepSeek, released October 2023. Its published size is 6.74B parameters and a 16k-token context window. Licence: deepseek.

Catalogue identity notes

Maker repo deepseek-ai/deepseek-coder-6.7b-base @ce2207a: licence tag other (deepseek-license). 6,740,512,768 parameters from the Hugging Face safetensors metadata (dense, so active = total). Released 2023-10: month the maker's Hugging Face repo was created (createdAt 2023-10-23). DeepSeek-Coder v1 base model. Ollama tags deepseek-coder:6.7b-base-q4_K_M, deepseek-coder:6.7b-base-q8_0 (sizes as listed on the tags page, observed 2026-09-29).

39.1±0.6#544 of 719 LLMsfrom 12 benchmark results across 5 sources

$ how it compares

The LLMs ranked either side of DeepSeek-Coder 6.7B Base, out of 828 with a score, then more from DeepSeek.

#modelscorewhereRAMreleased
634Mistral 7B Instruct v0.3 Mistral39±1local~5.8GB2024-05
635Vicuna 33B v1.3 LMSYS39±1local~23GB2023-06
636DeepSeek-Coder 6.7B Base DeepSeek39±1local~5.8GB2023-10
637Starling-LM 7B alpha UC Berkeley (NEST)39±1local~6.1GB2023-11
638Granite 8B Code Base (4K) IBM39±1local~6.6GB2024-04
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

coding▁▁▁▃▆██▇▆▄▃▁ #385 of 458 · 87% of its score · 7 tests

Beats DeepSeek R1 Distill Qwen 7B, Qwen3.5 2B, DeepSeek R1 Distill Llama 8B here, though they rank higher overall. Trails Qwen2.5 1.5B Instruct, WizardLM 13B V1.2, Falcon3 3B Instruct here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
HumanEval +247.6% ✎▁·▁▁▃▄▆▅█▃▁▁ beats 32/99easy · avg 79%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 54%sharp · at its level · several runs
HumanEval+39.6% ✎▁·▁▁▁▃▄▄▄█▃▁ beats 30/99easy · avg 74%▇▇▇▇▇▇▇▇ 20%sharp · at its level
MBPP+58.7% ✎▁···▁▁▂▅▇█▆▂ beats 34/75easy · avg 68%▇ 3.7%at its level
MBPP70.2% ✎▄▄▄██▄·▄█·▄▄ beats 11/12easy · avg 82%▇ 3.4%at its level
DS-100028.9% ✎▄▄·▄██▄█·█·▄ beats 9/12medium · avg 52%▇ 3.3%at its level
CRUXEval-I41.9% ✎▁··▁▅▂█▇▁··▁ beats 13/22medium · avg 54%▇ 1.1%at its level
CRUXEval-O43.5% ✎▁·▂▃▁█▁▂···▁ beats 17/22medium · avg 53%▇ 1.1%blunt · at its level
math▁▂▃▅▆▅▄█▄▁▁▁ #353 of 398 * · 2.2% of its score · 1 test

Beats Yi 34B Chat, Llama 3 8B Instruct, Command R (03-2024) here, though they rank higher overall. Trails Gemma 3 1B, Qwen3 0.6B, Tulu 2 DPO 7B here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
GSM8K41.9% ✎▄▄▄·▄█·██·█▄ beats 9/12easy · avg 75%▇ 2.2%sharp · at its level
other tests2 not in a task · 11% of its score
testresultvs the others in its rundifficultycountswhy
EvoEval (overall)32.9% ✎▂▂▃▄▄█▁▄▅▃▁▃ beats 19/42medium · avg 55%▇▇▇ 8.0%at its level
BigCodeBench (Complete)41.8% ✎▂▁▁▄▂▅▇█▆▂▂▂ beats 31/58medium · avg 50%▇ 3.2%at its level

$ 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 6.7B Base · needs ~4.1 GB · from huggingface.co/TheBloke
    chat in the terminal
    llama cli -hf TheBloke/deepseek-coder-6.7B-base-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf TheBloke/deepseek-coder-6.7B-base-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 6.7B Base · needs ~4.1 GB · from huggingface.co/TheBloke

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/TheBloke/deepseek-coder-6.7B-base-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 6.7B Base · needs ~4.1 GB · from ollama.com
    chat in the terminal
    ollama run deepseek-coder:6.7b-base-q4_K_M

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

$ how the score was worked out · EvalPlus / EvalPerf published… 53%, Top Leaderboard Ranking = Top… 23%, StarCoder 2 and The Stack v2:… 19% and 2 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
EvalPlus / EvalPerf published evaluation runs3▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 53%Apache-2.0
Top Leaderboard Ranking = Top Coding Proficiency, Always? EvoEval: Evolving Coding Benchmarks via LLM — authors' own runs2▇▇▇▇▇▇▇ 23%CC BY 4.0
StarCoder 2 and The Stack v2: The Next Generation — authors' own runs4▇▇▇▇▇▇ 19%CC BY-SA 4.0
BigCodeBench paper (Zhuo et al., ICLR 2025), arXiv 2406.15877v41▇ 3%CC BY 4.0
CRUXEval published evaluation results (authors' generations and scores)2▇ 2%MIT
prior · pull toward the middle▇ 0%
weighted average= 39.1

$ 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