Radamanthys11/Kimi-K2.5-DFlash-GGUF overview
Kimi K2.5 DFlash GGUF GGUF made to use in ikawrakow/ik llama.cpp https://github.com/ikawrakow/ik llama.cpp , currently for PR 1970 https://github.com/ikawrakow…
Runs locally from ~1.90 GB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
Model Details
| Model ID | Radamanthys11/Kimi-K2.5-DFlash-GGUF |
|---|---|
| Author | Radamanthys11 |
| Pipeline | text-generation |
| License | — |
| Base model | z-lab/Kimi-K2.5-DFlash |
| Last modified | 2026-06-15T02:02:31.000Z |
Model README
---
base_model: z-lab/Kimi-K2.5-DFlash
tags:
- gguf
- dflash
- ik_llama.cpp
library_name: gguf
pipeline_tag: text-generation
---
Kimi K2.5 DFlash GGUF
GGUF made to use in ikawrakow/ik_llama.cpp, currently for PR #1970. The small quantizations delivered here are made for test purposes; feel free to create your own quantization.
Derived from the safetensors DFlash draft model z-lab/Kimi-K2.5-DFlash.
Compatible target model
Kimi-K2.5.gguf- Mainly tested with IQ1_KT.
Files
| File | Quant | Size |
|---|---|---|
| kimi-k2-5-dflash-F16.gguf | F16 | 6.5 GB |
| kimi-k2-5-dflash-Q8_0.gguf | Q8_0 | 3.5 GB |
| kimi-k2-5-dflash-Q4_K_M.gguf | Q4_K_M | 1.9 GB |
Usage
./build/bin/llama-server \
--model <target.gguf> \
--model-draft <draft.gguf> \
--spec-type dflash:n_max=<N>,cross_ctx=<N> ...Notes
- This repo contains DFlash draft models, not a standalone instruct model.
- Use it with the matching target family listed above.
Q4_K_MandQ8_0are small test-oriented quants; create your own quant if you need a different tradeoff.
Run Radamanthys11/Kimi-K2.5-DFlash-GGUF with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models