Kimi-K3-DSpark vs DeepSeek-R1-0528-Qwen3-8B-MLX-8bit
Specs, VRAM requirements and download trends — updated 21 September 2026.
lmstudio-community · B
Params
2.3 B
Context
131,072
30d
271 K
mit · commercial OK
Specification comparison
Differences are highlighted; identical values are muted.
| Specification | Kimi-K3-DSpark | DeepSeek-R1-0528-Qwen3-8B-MLX-8bit |
|---|---|---|
| Parameters | 2.3 B | 2.3 B |
| Architecture | DSparkDraftModel | Qwen3ForCausalLM |
| Context length | 1,048,576 | 131,072 |
| Licence | — | mit |
| Commercial use | Unknown | Allowed |
| Languages | — | — |
| Downloads 30d | 3,658,082 | 270,667 |
| Downloads all time | 6.5 M | 5.3 M |
| Quantizations on the Hub | SAFETENSORS | SAFETENSORS |
| Gated | No | No |
| First seen on the Hub | 2026-07-27 | 2025-05-29 |
Download trend
Daily snapshots, last 51 days (2 Aug – 21 Sep)
3.7 M168 K
Kimi-K3-DSpark
DeepSeek-R1-0528-Qwen3-8B-MLX-8bit
A overtook B on 3 Aug
VRAM side by side
On RTX 4090 · 24 GB · 8K context unless noted
DeepSeek-R1-0528-Qwen3-8B-MLX-8bit · fp16 @ 128K ctx
15.6 GB / 24 GB
Kimi-K3-DSpark · fp16 @ 1024K ctx
48.6 GB / 24 GB