decart-ai / text-generation updated 1 month ago

Kimi-K2.7-Code-NVFP4

Name: Kimi-K2.7-Code-NVFP4
Author: decart-ai

NVFP4 quantized version of moonshotai/Kimi-K2.7-Code, quantized with NVIDIA Model Optimizer. The routed-expert linear layers are quantized to NVFP4 (4-bit float, block size 16) with an FP8 KV cache; attention (MLA), shared experts, the layer-0 dense MLP, lmhead, and the vision tower / mmprojector remain BF16 — the same precision split as nvidia/Kimi-K2.6-NVFP4. Ready for inference with vLLM on NVIDIA Blackwell.

Params

—

Context

—

Downloads 30d

194 K

Likes

Licence: other Not gated SAFETENSORS View on Hugging Face ↗

Download history

daily snapshots · 10 days

194 K194 K

Jul 28Jul 31Aug 3Aug 6

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
model.safetensors	fp16	595.2 GB	655.2 GB	❌ Won’t fit

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

~ · curl · api/v1

$ curl -s https://aimodelscomparison.com/api/v1/models/kimi-k2-7-code-nvfp4-decart-ai
{
  "hf_id": "decart-ai/Kimi-K2.7-Code-NVFP4",
  "params_b": null,
  "context_length": null,
  "license": { "id": "other", "commercial": "unknown" },
  "downloads_30d": 193713,
  "vram_estimates": [
    { "quant": "fp16", "gb": 655.2 }
  ],
  "updated_at": "2026-07-28T18:08:14Z"
}

est. VRAM —on RTX 4090 · 24 GBJSON API →

Specifications

Architecture: KimiK25ForConditionalGeneration
Licence: other
First seen on the Hub: 2026-06-19
Training datasets: undisclosed
Added to our catalog: 2026-07-28

Appears in

Best open coding models Commercial-use text models Models that fit in 12 GB