[GH-ISSUE #8523] Deepseek-R1 quants #31255

Closed
opened 2026-04-22 11:32:49 -05:00 by GiteaMirror · 2 comments
Owner

Originally created by @stutteringp0et on GitHub (Jan 21, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/8523

Many older models are offered with various q4-8 and fp16 quants, but lately we get 4, 8, and fp16 only

I'd really love to see a Q6_K quant of these.

Originally created by @stutteringp0et on GitHub (Jan 21, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/8523 Many older models are offered with various q4-8 and fp16 quants, but lately we get 4, 8, and fp16 only I'd really love to see a Q6_K quant of these.
GiteaMirror added the model label 2026-04-22 11:32:49 -05:00
Author
Owner

@rick-github commented on GitHub (Jan 22, 2025):

Until other quants are made available, if you are willing to pull the fp16 quant, you can make your own q6_k:

$ ollama pull deepseek-r1:1.5b-qwen-distill-fp16
$ ollama show --modelfile deepseek-r1:1.5b-qwen-distill-fp16 > Modelfile
$ ollama create --quantize Q6_K deepseek-r1:1.5b-qwen-distill-Q6_K

$ ollama show deepseek-r1:1.5b-qwen-distill-Q6_K
  Model
    architecture        qwen2     
    parameters          1.8B      
    context length      131072    
    embedding length    1536      
    quantization        Q6_K      

  Parameters
    stop    "<|begin▁of▁sentence|>"    
    stop    "<|end▁of▁sentence|>"      
    stop    "<|User|>"                 
    stop    "<|Assistant|>"            

  License
    MIT License                    
    Copyright (c) 2023 DeepSeek    

$ ollama run deepseek-r1:1.5b-qwen-distill-Q6_K
>>> how many 'r's in `strawberry`?
<think>
To determine how many times the letter 'r' appears in the word "strawberry," I'll start by listing all the characters of the word.

Next, I'll scan each character to check if it matches 'r'. 

Finally, I'll count each occurrence of 'r' that meets this condition.
</think>

To determine how many times the letter **'r'** appears in the word "strawberry," let's break down the process step by step.

### Step 1: Identify All Characters
The word "strawberry" consists of the following characters:
\[ s,\, t,\, r,\, a,\, w,\, b,\, e,\, r,\, r,\, y \]

### Step 2: Count the Occurrences of 'r'
Now, let's count how many times **'r'** appears in this list:

1. **s:** Not an 'r.'
2. **t:** Not an 'r.'
3. **r:** This is one 'r.'
4. **a:** Not an 'r.'
5. **w:** Not an 'r.'
6. **b:** Not an 'r.'
7. **e:** Not an 'r.'
8. **r:** This is the second 'r.'
9. **r:** This is the third 'r.'
10. **y:** Not an 'r.'

### Step 3: Total Count
By counting each occurrence, we find that there are:

\[
\boxed{3}
\]

**Answer:** The letter **'r'** appears **3 times** in the word "strawberry."
<!-- gh-comment-id:2606005229 --> @rick-github commented on GitHub (Jan 22, 2025): Until other quants are made available, if you are willing to pull the fp16 quant, you can make your own q6_k: ```console $ ollama pull deepseek-r1:1.5b-qwen-distill-fp16 $ ollama show --modelfile deepseek-r1:1.5b-qwen-distill-fp16 > Modelfile $ ollama create --quantize Q6_K deepseek-r1:1.5b-qwen-distill-Q6_K $ ollama show deepseek-r1:1.5b-qwen-distill-Q6_K Model architecture qwen2 parameters 1.8B context length 131072 embedding length 1536 quantization Q6_K Parameters stop "<|begin▁of▁sentence|>" stop "<|end▁of▁sentence|>" stop "<|User|>" stop "<|Assistant|>" License MIT License Copyright (c) 2023 DeepSeek $ ollama run deepseek-r1:1.5b-qwen-distill-Q6_K >>> how many 'r's in `strawberry`? <think> To determine how many times the letter 'r' appears in the word "strawberry," I'll start by listing all the characters of the word. Next, I'll scan each character to check if it matches 'r'. Finally, I'll count each occurrence of 'r' that meets this condition. </think> To determine how many times the letter **'r'** appears in the word "strawberry," let's break down the process step by step. ### Step 1: Identify All Characters The word "strawberry" consists of the following characters: \[ s,\, t,\, r,\, a,\, w,\, b,\, e,\, r,\, r,\, y \] ### Step 2: Count the Occurrences of 'r' Now, let's count how many times **'r'** appears in this list: 1. **s:** Not an 'r.' 2. **t:** Not an 'r.' 3. **r:** This is one 'r.' 4. **a:** Not an 'r.' 5. **w:** Not an 'r.' 6. **b:** Not an 'r.' 7. **e:** Not an 'r.' 8. **r:** This is the second 'r.' 9. **r:** This is the third 'r.' 10. **y:** Not an 'r.' ### Step 3: Total Count By counting each occurrence, we find that there are: \[ \boxed{3} \] **Answer:** The letter **'r'** appears **3 times** in the word "strawberry." ```
Author
Owner

@stutteringp0et commented on GitHub (Jan 28, 2025):

This is excellent information. Thank you for posting it @rick-github

Saved for future reference!

<!-- gh-comment-id:2619667229 --> @stutteringp0et commented on GitHub (Jan 28, 2025): This is excellent information. Thank you for posting it @rick-github Saved for future reference!
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#31255