DeepSeek-R1-0528-Qwen3-8B-GGUF
Learn how to run DeepSeek-R1-0528 correctly - Read our Guide. See our collection for all versions of R1 including GGUF, 4-bit & 16-bit formats. Unsloth Dynamic 2.0 achieves superior accuracy & outperforms other leading quants. 🐋 DeepSeek-R1-0528-Qwen3-8B Usage Guidelines
Params
8.0 B
Context
131,072
Downloads 30d
161 K
Likes
484
Download history
daily snapshots · 15 days161 K139 K
Oct 27Nov 1Nov 6Nov 10
Can you run it?
Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.
| File | Quant | Size | Est. VRAM | Verdict on RTX 4090 · 24 GB |
|---|---|---|---|---|
| DeepSeek-R1-0528-Qwen3-8B-Q2_K.gguf | Q2_K | 3.3 GB | 5.3 GB | ✅ Runs comfortably |
| DeepSeek-R1-0528-Qwen3-8B-Q2_K_L.gguf | Q2_K_L | 3.4 GB | 5.5 GB | ✅ Runs comfortably |
| DeepSeek-R1-0528-Qwen3-8B-Q3_K_S.gguf | Q3_K_S | 3.8 GB | 5.8 GB | ✅ Runs comfortably |
| DeepSeek-R1-0528-Qwen3-8B-Q3_K_M.gguf | Q3_K_M | 4.1 GB | 6.2 GB | ✅ Runs comfortably |
| DeepSeek-R1-0528-Qwen3-8B-IQ4_XS.gguf | IQ4_XS | 4.6 GB | 6.7 GB | ✅ Runs comfortably |
| DeepSeek-R1-0528-Qwen3-8B-IQ4_NL.gguf | IQ4_NL | 4.8 GB | 7.0 GB | ✅ Runs comfortably |
| DeepSeek-R1-0528-Qwen3-8B-Q4_0.gguf | Q4_0 | 4.8 GB | 7.0 GB | ✅ Runs comfortably |
| DeepSeek-R1-0528-Qwen3-8B-BF16.gguf | GGUF | 16.4 GB | 19.7 GB | ⚠️ Tight — reduce context |
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.
Run it
copy-paste, exact tags checked against the Hub$ ollama run deepseek-r1-0528-qwen3-8b-gguf # pin the quantization explicitly $ ollama run deepseek-r1-0528-qwen3-8b-gguf-q4_0
$ huggingface-cli download unsloth/DeepSeek-R1-0528-Qwen3-8B-GGUF-GGUF \
DeepSeek-R1-0528-Qwen3-8B-Q4_0.gguf --local-dir .
$ llama-cli -m DeepSeek-R1-0528-Qwen3-8B-Q4_0.gguf \
-c 8192 -ngl 99 -t 8 --color
$ curl -s https://aimodelscomparison.com/api/v1/models/deepseek-r1-0528-qwen3-8b-gguf
{
"hf_id": "unsloth/DeepSeek-R1-0528-Qwen3-8B-GGUF",
"params_b": 8.00,
"context_length": 131072,
"license": { "id": "mit", "commercial": "yes" },
"downloads_30d": 160771,
"vram_estimates": [
{ "quant": "GGUF", "gb": 19.7 },
{ "quant": "IQ4_NL", "gb": 7.0 }
],
"updated_at": "2026-10-27T01:00:24Z"
}
Specifications
- Architecture
- Qwen3ForCausalLM
- Parameters
- 8.0 B
- Context length
- 131,072
- Vocabulary
- 151,936
- Layers / heads
- 36 / 32
- Licence
- mit
- First seen on the Hub
- 2025-05-29
- Base model
- DeepSeek-R1-0528-Qwen3-8B
- Training datasets
- undisclosed
- Added to our catalog
- 2026-10-27
Family
Base model and the most-downloaded derivatives in the catalog.
Compare with any text-generation model