[whatmodel.sh](../../README.md) / DeepSeek R1 Distill Qwen 1.5B
# DeepSeek R1 Distill Qwen 1.5B
DeepSeek · released 2025-01 · 1.78B params · mit
**score 36.7** ±1.7 · #612 of 732 · from 16 benchmark results across 7 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks DeepSeek R1 Distill Qwen 1.5B among every model on that task or test, worst to best.
task vs the others rank counts tests
coding ▁●▁▃▅▇█▇▅▄▃▁ #465 of 469 28% 5
math ▁▁▃▅█●▄█▄▂▁▁ #254 of 431 32% 4
coding · #465 of 469 · 28% of its score
Trails [GPT-Neo 2.7B](gpt-neo-2-7b.md), [Phi-1 (1.3B)](phi-1.md), [Pythia 6.9B](pythia-6-9b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
LiveCodeBench (version unspecifie… 16.9% ●····▄▄··▄█▄ beats 0/6 medium · avg 42% 15% typical
CodeReviewQA: solution identifica… 0.0% ●··▁▂▂▃▂█▃▄▂ beats 2/47 medium · avg 51% 4.5% sharp · noisy
CodeReviewQA: automated code refi… 2.2% ●▁▁▁▁▁▂▂▃█▅▄ beats 0/47 hard · avg 32% 4.5% typical
CodeReviewQA: change localisation… 0.0% ●▂▁▁▂▂▂▃▅▅▅▇ beats 3/47 hard · avg 33% 2.9% sharp · too hard for it · noisy
CodeReviewQA: change type recogni… 0.0% ●▁▁······▁▃█ beats 1/47 easy · avg 67% 0.3% blunt · at its level · noisy
math · #254 of 431 · 32% of its score
Beats [Command A (03-2025)](command-a.md), [GPT-4o (2024-05-13)](gpt-4o-2024-05-13.md), [GPT-4o (2024-11-20)](gpt-4o-2024-11-20.md) here, though they rank higher overall.
test result vs the others in its run difficulty counts why
MATH-500 (maker) 83.9% ●·▄▄·▄▄█···▄ beats 0/7 very easy · avg 91% 18% typical
THINK-Bench final-answer accuracy… 62.9% ●█·▄····▄▄·█ beats 0/7 easy · avg 76% 8.1% at its level
AIME 2024 (maker) 28.9% ●·▃·▃▃···█·▃ beats 0/7 medium · avg 50% 5.1% sharp · noisy
OTIS mock AIME 2024–2025 21.4% ▃▃▄▃●▃▇▇█▃▂▅ beats 59/197 very hard · avg 7.2% 0.3% sharp · too hard for it · noisy
other tests (6)
test result vs the others in its run difficulty counts why
BRIDGE clinical text, zero-shot a… 14.3% ▁▁●▁▁▂▂▃▄▆█▃ beats 3/97 hard · avg 31% 13% blunt
GPQA Diamond 33.6% ▁▁▁▁●▄▆▄▅▇█▅ beats 24/211 hard · avg 39% 12% sharp · too hard for it · several runs
Open PT-LLM Leaderboard average (… 25.3% ●▁▁▁▁▁▁▁▂█▅▁ beats 26/276 medium · avg 72% 6.2% at its level · noisy
EuroEval Life in the UK (English … 29.9% ▁▁▁●▁▁▁▂▄█▄▃ beats 6/146 easy · avg 82% 5.0% at its level · noisy
EuroEval HellaSwag (English), tes… 32.1% ▃·▁▁●▁▂▅█▇▄▁ beats 14/100 medium · avg 67% 4.1% sharp · noisy
Chess Puzzles (Epoch) 0.0% ●▁▁▁▁▂▃▃▃▂▁▁ beats 25/137 very hard · avg 5.8% 0.1% sharp · too hard for it · noisy
source results share of the score
deepseek-ai/DeepSeek-R1 model card (sub… 4 44% ▇▇▇▇▇
BRIDGE Medical Leaderboard 1 13% ▇▇
CodeReviewQA: The Code Review Comprehen… 4 12% ▇
EuroEval leaderboards 2 9% ▇
THINK-Bench: Evaluating Thinking Effici… 1 8% ▇
Epoch AI 3 7% ▇
Open Portuguese LLM Leaderboard 1 6% ▇
prior: pull toward the middle 1% ▇
weighted average = 36.7
Licences: deepseek-ai/DeepSeek-R1 model card ([mit](../../licenses/deepseek-ai--DeepSeek-R1.txt)) · BRIDGE Medical Leaderboard ([MIT](../../licenses/bridge-medical-leaderboard-MIT.txt)) · CodeReviewQA: The Code Review Comprehension Assessment for Large Language Models ([CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/)) · EuroEval leaderboards ([MIT](https://github.com/EuroEval/leaderboards/blob/1a77f1d2f8ba1ba05ab73287fb090d76bf41b2e0/LICENSE)) · THINK-Bench: Evaluating Thinking Efficiency and Chain-of-Thought Quality of Large Reasoning Models ([CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/)) · Epoch AI ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Open Portuguese LLM Leaderboard ([Apache-2.0](../../licenses/hf-leaderboard-spaces-Apache-2.0.txt))
## How to run it
### llama.cpp (recommended)
1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md)
```sh
# macOS / Linux
curl -LsSf https://llama.app/install.sh | sh
# Windows
irm https://llama.app/install.ps1 | iex
```
2. Run DeepSeek R1 Distill Qwen 1.5B (Q4_K_M, needs ~1.1 GB, from [huggingface.co/lmstudio-community](https://huggingface.co/lmstudio-community/DeepSeek-R1-Distill-Qwen-1.5B-GGUF/tree/29f63ea6272ffa55e5e93632a33a3e2ec95dbd05))
```sh
llama cli -hf lmstudio-community/DeepSeek-R1-Distill-Qwen-1.5B-GGUF:Q4_K_M
```
3. Or start a local server and open the address it prints (http://localhost:9931)
```sh
llama serve -hf lmstudio-community/DeepSeek-R1-Distill-Qwen-1.5B-GGUF:Q4_K_M
```
Other versions: Q8_0 (~1.9 GB) `llama cli -hf lmstudio-community/DeepSeek-R1-Distill-Qwen-1.5B-GGUF:Q8_0`
### LM Studio
1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download)
2. Run DeepSeek R1 Distill Qwen 1.5B (Q4_K_M, needs ~1.1 GB, from [huggingface.co/lmstudio-community](https://huggingface.co/lmstudio-community/DeepSeek-R1-Distill-Qwen-1.5B-GGUF/tree/29f63ea6272ffa55e5e93632a33a3e2ec95dbd05))
```sh
lms get https://huggingface.co/lmstudio-community/DeepSeek-R1-Distill-Qwen-1.5B-GGUF@Q4_K_M -y
```
3. Then run lms ls to see its name, and lms chat followed by that name
Other versions: Q8_0 (~1.9 GB) `lms get https://huggingface.co/lmstudio-community/DeepSeek-R1-Distill-Qwen-1.5B-GGUF@Q8_0 -y` · 4bit (~1 GB, Mac only) `lms get https://huggingface.co/mlx-community/DeepSeek-R1-Distill-Qwen-1.5B-4bit -y` · 8bit (~1.9 GB, Mac only) `lms get https://huggingface.co/mlx-community/DeepSeek-R1-Distill-Qwen-1.5B-8bit -y`
### Ollama
1. Install Ollama. [Download it (macOS / Windows)](https://ollama.com/download)
```sh
# Linux
curl -fsSL https://ollama.com/install.sh | sh
```
2. Run DeepSeek R1 Distill Qwen 1.5B (Q4_K_M, needs ~1.57 GB, from [ollama.com](https://ollama.com/library/deepseek-r1/tags))
```sh
ollama run deepseek-r1:1.5b
```
Other versions: Q8_0 (~2.64 GB) `ollama run deepseek-r1:1.5b-qwen-distill-q8_0`
### MLX (Mac)
1. Install MLX (Mac). Apple silicon Macs only, text only.
```sh
# with uv
uv tool install mlx-lm
# or
pip install mlx-lm
```
2. Run DeepSeek R1 Distill Qwen 1.5B (4bit, needs ~1 GB, Mac only, from [huggingface.co/mlx-community](https://huggingface.co/mlx-community/DeepSeek-R1-Distill-Qwen-1.5B-4bit/tree/933185be1b8f81d9a21dcfa15ff73470d3545240))
```sh
mlx_lm.chat --model mlx-community/DeepSeek-R1-Distill-Qwen-1.5B-4bit
```
3. Or start an OpenAI-compatible server at http://localhost:8080/v1
```sh
mlx_lm.server --model mlx-community/DeepSeek-R1-Distill-Qwen-1.5B-4bit
```
Other versions: 8bit (~1.9 GB, Mac only) `mlx_lm.chat --model mlx-community/DeepSeek-R1-Distill-Qwen-1.5B-8bit`
Model repository and documentation: [huggingface.co/deepseek-ai](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B)
## Quantization evidence
How much of the full-precision result a quantized version kept, measured within one study. Rows from the same study share a table and aren't independent. They don't describe every download of that precision.
precision test full → quantized kept
FP8 W8A8 dynamic (FP8 weights… GPQA Diamond pass@1 0.354 → 0.359 101.6%
FP8 W8A8 dynamic (FP8 weights… IFEval 0-shot instruction-level s… 0.346 → 0.349 100.8%
FP8 W8A8 dynamic (FP8 weights… Reasoning average of 3 tasks (AIM… 0.5 → 0.502 100.3%
FP8 W8A8 dynamic (FP8 weights… MATH-500 pass@1 0.847 → 0.847 100.1%
FP8 W8A8 dynamic (FP8 weights… OpenLLM v1 average of 6 tasks (AR… 0.48 → 0.478 99.6%
FP8 W8A8 dynamic (FP8 weights… MMLU 5-shot acc (lm-evaluation-ha… 0.374 → 0.372 99.6%
FP8 W8A8 dynamic (FP8 weights… AIME 2024 pass@1 0.3 → 0.298 99.3%
FP8 W8A8 dynamic (FP8 weights… GSM8K 5-shot strict-match (lm-eva… 0.7 → 0.69 98.6%
INT4 W4A16 (weights only, sym… OpenLLM v1 average of 6 tasks (AR… 0.48 → 0.476 99.1%
INT4 W4A16 (weights only, sym… MMLU 5-shot acc (lm-evaluation-ha… 0.374 → 0.37 98.9%
INT4 W4A16 (weights only, sym… GSM8K 5-shot strict-match (lm-eva… 0.7 → 0.68 97.2%
INT4 W4A16 (weights only, sym… GPQA Diamond pass@1 0.354 → 0.332 93.8%
INT4 W4A16 (weights only, sym… Reasoning average of 3 tasks (AIM… 0.5 → 0.468 93.5%
INT4 W4A16 (weights only, sym… AIME 2024 pass@1 0.3 → 0.246 82.0%
Each kept value links to its source.
[interactive version](https://whatmodel.sh/model/deepseek-r1-distill-qwen-1-5b) · [JSON](https://whatmodel.sh/api?format=json&all=1)