GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
MirilAI
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
MirilAI
Model Tree
MirilAI/Miril-Drone-2B-1
Input Modalities
Output Modalities
Supported Functionality
Use transformers>=5.12.1. This model retains Gemma 4 E2B's declared shared-KV layout; language layers 15 through 34 intentionally reuse key/value states and do not carry separate k_proj, v_proj, or k_norm tensors.
bash
python eval_generate.py \--model_id MirilAI/Miril-Drone-2B-1-bnb4 \--processor_id MirilAI/Miril-Drone-2B-1-bnb4 \--jsonl eval.jsonl \--image_root images \--out_jsonl predictions.jsonl \--batch_size 1 \--max_new_tokens 256
MIRIL_VARIANT_PROFILE_PENDING
MIRIL_RELEASE_METRICS_PENDING
The merged and quantized variants are evaluated on identical held-out cases. Automated scores are regression signals, not safety certification.
Apache License 2.0. See LICENSE and NOTICE.