Qwen3-1.7B-Base VRAM & Memory Requirements
Specific checkpoints with sourced architecture and weight files. GGUF Q4_K_M (mradermacher): 2.03 GiB–2.49 GiB at 8,192 tokens, 1 sequence, FP16 KV, llama.cpp.
This result is an estimate, not a measured trace.Not verifiedFile-backed weights
Publisher and provenance
- PublisherQwen
- Hugging Face
Qwen/Qwen3-1.7B-Base - Revision
ea980cb0a6c2ae4b936e82123acc929f1cec04c1 - Total parameters1,720,574,976
- Layers28
- Query heads16
- KV heads8 (gqa)
- Head dimension128
- Recorded context capability32,768
- Completenesspartial · full
Memory layers in this estimate
Artifact bytes, raw KV, engine extras, and total demand are separate. This is not a runtime trace.
- Weight artifact
- 1.03 GiB (1,107,409,024 bytes)
- Raw KV cache
- 0.88 GiB (939,524,096 bytes)
- Engine extras / workspace
- 0.13 GiB–0.58 GiB
- Total estimated demand
- 2.03 GiB–2.49 GiB
Published artifacts
| Quantization | Selected files | Bytes | Engine | Evidence |
|---|---|---|---|---|
| GGUF Q4_K_M (mradermacher) gguf-q4_k_m | Qwen3-1.7B-Base.Q4_K_M.gguf | 1,107,409,024 (1.03 GiB / 1.11 GB) | llama.cpp, ollama, unspecified | file-metadata |
| GGUF Q5_K_M (mradermacher) gguf-q5_k_m | Qwen3-1.7B-Base.Q5_K_M.gguf | 1,257,879,680 (1.17 GiB / 1.26 GB) | llama.cpp, ollama, unspecified | file-metadata |
| GGUF Q6_K (mradermacher) gguf-q6_k | Qwen3-1.7B-Base.Q6_K.gguf | 1,417,754,752 (1.32 GiB / 1.42 GB) | llama.cpp, ollama, unspecified | file-metadata |
| GGUF Q8_0 (mradermacher) gguf-q8_0 | Qwen3-1.7B-Base.Q8_0.gguf | 1,834,426,496 (1.71 GiB / 1.83 GB) | llama.cpp, ollama, unspecified | file-metadata |
| INT4 theoretical | No published file | 860,287,488 (theoretical) | — | theoretical |
Memory by quantization and context
Concurrent sequences = 1. KV defaults to FP16. The range includes documented runtime extras for the listed engine. These figures use the same core as the interactive calculator.
| Quantization | 2,048 tokens | 8,192 tokens | 32,768 tokens |
|---|---|---|---|
| GGUF Q4_K_M (mradermacher) | 1.38 GiB–1.78 GiB | 2.03 GiB–2.49 GiB | 4.66 GiB–5.30 GiB |
| GGUF Q5_K_M (mradermacher) | 1.52 GiB–1.92 GiB | 2.17 GiB–2.63 GiB | 4.80 GiB–5.44 GiB |
| GGUF Q6_K (mradermacher) | 1.66 GiB–2.07 GiB | 2.32 GiB–2.78 GiB | 4.95 GiB–5.59 GiB |
| GGUF Q8_0 (mradermacher) | 2.05 GiB–2.47 GiB | 2.71 GiB–3.18 GiB | 5.33 GiB–5.99 GiB |
| INT4 theoretical Theoretical rows are not downloadable artifacts. | 1.14 GiB–1.86 GiB | 1.80 GiB–2.68 GiB | 4.43 GiB–5.91 GiB |
Raw KV contribution
Standard GQA/MHA raw KV is 2 × layers × kvHeads × headDim × tokens × sequences × bytes per element. This page uses KV heads (8), not query heads (16). Using query heads would overstate this cache by 2×.
- 8,192 tokens, 1 sequence, FP16: 939,524,096 bytes (0.8750 GiB)
- 32,768 tokens: 3,758,096,384 bytes (3.5000 GiB)
Related pages
Sources
- Qwen3-1.7B-Base model card
- config.json — revision ea980cb0a6c2ae4b936e82123acc929f1cec04c1
- Artifact repo mradermacher/Qwen3-1.7B-Base-GGUF
Assumptions that affect this result
- Requested context is within the recorded native window of 32768 tokens.
- Full-resident weight estimates use total parameters, including inactive MoE experts.
- Weight bytes come from the selected weight files only, not from every file in the repository.
- Download file size is not peak GPU memory. Runtime layout, allocator padding, and KV are extra.
- Counted files: Qwen3-1.7B-Base.Q4_K_M.gguf.
- Context budget is the number of cached tokens per sequence (prompt plus reserved generation).
- Concurrent sequences share one model replica; weights are not multiplied by sequence count.
- KV dtype is independent of weight quantization unless you change it.
Calculator
Loading catalog snapshot…
Search or pick a catalog checkpoint to see memory demand.
Search or pick a catalog checkpoint to see memory demand.