Simon Willison's Weblog

That model was trained in part using their unreleased R1 "reasoning" design. Today they have actually launched R1 itself, along with an entire family of brand-new designs obtained from that base.

There's a great deal of stuff in the new release.

DeepSeek-R1-Zero appears to be the base design. It's over 650GB in size and, like the majority of their other releases, is under a clean MIT license. DeepSeek caution that "DeepSeek-R1-Zero encounters difficulties such as unlimited repetition, poor readability, and language mixing." ... so they also released:

DeepSeek-R1-which "incorporates cold-start data before RL" and "attains efficiency similar to OpenAI-o1 across mathematics, code, and reasoning jobs". That one is likewise MIT licensed, and is a similar size.

I don't have the ability to run designs bigger than about 50GB (I have an M2 with 64GB of RAM), so neither of these two models are something I can easily have fun with myself. That's where the brand-new distilled models are available in.

To support the research study neighborhood, we have open-sourced DeepSeek-R1-Zero, DeepSeek-R1, and six thick designs distilled from DeepSeek-R1 based upon Llama and Qwen.

This is a fascinating flex! They have actually models based upon Qwen 2.5 (14B, 32B, Math 1.5 B and Math 7B) and Llama 3 (Llama-3.1 8B and Llama 3.3 70B Instruct).

Weirdly those Llama models have an MIT license attached, which I'm uncertain works with the underlying Llama license. Qwen designs are Apache licensed so possibly MIT is OK?

Licensing aside, these distilled designs are remarkable monsters.

Running DeepSeek-R1-Distill-Llama-8B-GGUF

Quantized versions are already starting to appear. Up until now I've attempted just among those- unsloth/DeepSeek-R 1-Distill-Llama-8B-GGUF launched by Unsloth AI-and it's really enjoyable to play with.

I'm running it using the mix of Ollama, LLM and the llm-ollama plugin.

First I brought and ran the design using Ollama itself:

This downloads around 8.5 GB of model data and begins an interactive chat user interface.

Once the model has actually been brought, LLM can talk with it also. I choose using LLM for experiments due to the fact that it logs whatever to SQLite for later exploration.

I installed the plugin and ran it like this:

Each action begins with a ... pseudo-XML tag containing the chain of idea utilized to assist generate the reaction. For "a joke about a pelican and a walrus who run a tea space together" that started like this:

Okay, so I require to come up with a joke about a pelican and a walrus running a tea room together. Hmm, that's a fascinating mix. Let me think of how these two characters may interact in an amusing circumstance.

First, let's consider their characteristics. Pelicans are known for their long beaks and Webbed feet, typically seen near the beach or water. Walruses have huge teeth, thick fur, and they're normally found in colder climates, like icebergs or snowy locations. So, combining these two into a tea room setting is currently an amusing image.

[...]

It then believed for 20 paragraphs before outputting the joke! Here's the full records and here's the output at the end, after the tag:

The Joke:

A pelican and a walrus choose to open a quaint little tea space together. The walrus, with its huge size, has a hard time to discover comfy chairs, so it rests on the table by accident, knocking over the teapot. Meanwhile, the pelican, trying to assist, utilizes its beak to place dishes on the table, triggering a few spills.

After a series of humorous mishaps, the walrus searches for and says with a smile, "This isn't so fishy any longer." The pelican smirks and says, "Maybe not, but we do have a great deal of krill in our tea!"

Explanation:

- Setup: Pelican and Walrus run an unlikely tea space. Their sizes result in comedic turmoil.
- Punchline: "This isn't so fishy any longer" is a pun on the circumstance becoming less unreasonable. The follow-up joke introduces "krill," a play on words that connects into both marine life ("fish") and tea, making it funny while fitting the theme.

So yeah, the joke is horrible. But the procedure of arriving was such an intriguing insight into how these new models work.

This was a fairly little 8B model. I'm anticipating checking out the Llama 70B variation, which isn't yet available in a GGUF I can run with Ollama. Given the strength of Llama 3.3 70B-currently my favourite GPT-4 class design that I've run on my own machine-I have high expectations.

Update 21st January 2025: I got this quantized version of that Llama 3.3 70B R1 distilled model working like this-a 34GB download:

Can it draw a pelican?

I attempted my classic Generate an SVG of a pelican riding a bike prompt too. It did refrain from doing really well:

It aimed to me like it got the order of the elements wrong, so I followed up with:

the background wound up covering the remainder of the image

It thought some more and gave me this:

As with the earlier joke, the chain of thought in the records was far more intriguing than completion outcome.

Other methods to attempt DeepSeek-R1

If you wish to try the design out without installing anything at all you can do so utilizing chat.deepseek.com-you'll need to develop an account (check in with Google, use an email address or supply a Chinese +86 telephone number) and then pick the "DeepThink" option below the prompt input box.

DeepSeek provide the model through their API, using an OpenAI-imitating endpoint. You can access that by means of LLM by dropping this into your extra-openai-models. yaml setup file:

Then run llm secrets set deepseek and paste in your API secret, then utilize llm -m deepseek-reasoner 'prompt' to run prompts.

This won't reveal you the reasoning tokens, timeoftheworld.date sadly. Those are served up by the API (example here) but LLM doesn't yet have a method to show them.