DeepSeek V3

What DeepSeek V3 is for

DeepSeek V3 is trained for code. DeepSeek released it in 2024-12 with a single 671B (37B active MoE) size, 128K of context and a DeepSeek License licence. Context length matters more here than for chat, because a coding assistant that cannot see the whole file is guessing. The catalogue records its strength as frontier-class open-weights LLM (needs data-center GPUs), and it is the class of model an editor plugin such as Continue or Tabby points at.

Sizes and memory

Our catalogue does not carry a per-size memory breakdown for DeepSeek V3. What it records instead is a floor of 256 GB and a recommended 512 GB for the default configuration. Until a per-size figure exists, treat those two numbers as the whole guidance rather than reading a size table that would have to be invented.

Hardware that runs DeepSeek V3

Without a per-size memory figure there is no honest way to say which card holds DeepSeek V3. Its recorded floor of 256 GB rules out every consumer card in our hardware guide, which tops out well below that.

What people use it for

The catalogue lists 3 recorded uses for DeepSeek V3: research, frontier tasks, offline frontier inference. It is tagged MoE, general, code, math, frontier.

Where DeepSeek V3 sits in the DeepSeek family

Our catalogue holds 4 DeepSeek entries. Ordered by release date, DeepSeek V3 (2024-12) is the 3rd of them, and DeepSeek R1 (2025-01) is newer.

It ships under the DeepSeek License, which is neither a standard open-source licence nor a closed one. Read the terms before shipping anything commercial on top of it.

Context window: 128K

128K tokens is roughly 96,000 words at the usual ratio, which is a small repository or a whole book. Two cautions come with a window this size: the KV cache at full length can rival the weights for memory, and models rarely use the far end of their advertised context as well as they use the near end. Treat it as headroom, not as a promise.

How to run DeepSeek V3 locally

Our catalogue does not record registry tags, so look the current tag up in the Ollama library or on Hugging Face before pasting these. The commands below are the shape of the workflow, not a copy-and-paste recipe.

Option 1: Ollama (simplest)

# find the tag for DeepSeek V3 at ollama.com/library
ollama run <tag>

Option 2: Mullama (production)

mullama pull <tag>
mullama run <tag>

Option 3: llama.cpp (CLI)

# download a GGUF from Hugging Face, searching for "DeepSeek V3 GGUF"
./llama-cli -m model.Q4_K_M.gguf -p "Hello, AI!"

Option 4: Python with Mullama or llama-cpp-python

from mullama import Model, Context
model = Model.load("model.Q4_K_M.gguf", n_gpu_layers=99)
ctx = Context(model, n_ctx=4096)
print(ctx.generate("Hello, AI!", 256))

Sources

Model record from src/data/models.json (2026-06-29); card capacities from src/data/gpus.json (2026-06-29). Upstream: Curated from ollama.com/library + community benchmarks + paperswithcode + huggingface.