GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
MonumentalSystems
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
MonumentalSystems
Model Tree
Hcompany/Holo-3.1-0.8B
Input Modalities
Output Modalities
Supported Functionality
Atlas LoRA v0 applies the BF16 delta at attention k/v/o on the full-attention layers only. This adapter is shaped to match:
target_modules = ["k_proj", "v_proj", "o_proj"] — q_proj is gated on Holo and rejected; dense-FFN delta is not yet wired.layers_to_transform = [3, 7, 11, 15, 19, 23] — the 6 full-attention layers (the other 18 are Gated-DeltaNet/linear; Atlas hard-rejects a LoRA tensor on them).r = 32, lora_alpha = 64 (scale = α/r = 2.0), use_rslora = false, use_dora = false, bias = "none".Rank 32 (rather than a minimal 8) is used so the exact codeword digits survive the BF16-delta-on-NVFP4-base quantization mismatch when served by Atlas.
PEFT:
python
from transformers import AutoModelForCausalLM, AutoTokenizerfrom peft import PeftModelbase = AutoModelForCausalLM.from_pretrained("Hcompany/Holo-3.1-0.8B", trust_remote_code=True)model = PeftModel.from_pretrained(base, "MonumentalSystems/Holo-3.1-0.8B-lora-demo")
Atlas (requires the F32-adapter loader fix — adapter.rs F32→BF16 conversion):
bash
spark serve Hcompany/Holo-3.1-0.8B \--lora-adapter demo=MonumentalSystems/Holo-3.1-0.8B-lora-demo \--max-lora-rank 64
peft + transformers Trainer, ~10 epochs on 6 overfit Q→A pairs, r=32,
lora_alpha=64, lr 2e-4, bf16, ~4 min on a single NVIDIA GB10 (Grace-Blackwell,
aarch64, CUDA 13). This is a demo artifact, not a general-purpose fine-tune.