GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
GammaAGI
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
GammaAGI
Model Tree
Base
nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16
Adapter
this model
Input Modalities
Text
Output Modalities
Text
Supported Functionality
Dedicated Endpoints
Train a bounded, traceable EvoStream adapter on top of the accepted Nano scaffold using verified donor, code, reasoning, safety, and Omni contracts.
OK1.9799751.902021[0.804616, 1.098587, 0.804143, 0.52016, 0.938238, 0.901114, 0.638272, 1.048143, 2.29344, 2.541251, 1.279884, 1.117072, 1.022416, 0.987017, 1.022244, 1.236557, 1.609732, 4.831176, 1.078025, 1.070122, 1.658666, 1.126232, 4.182885, 3.996631, 4.371338, 4.395547, 3.608554, 2.560072, 3.105073, 3.924049, 3.763522, 3.00894, 1.26379, 3.034989, 3.009562, 3.600018, 3.906813, 4.105942, 0.622856, 1.089907, 1.647519, 1.348912, 1.098148, 0.780802, 0.805037, 1.1614, 1.385495, 1.201193, 1.08922, 0.877296, 0.74964, 1.054622, 0.700003, 0.453529, 0.833438, 0.77458, 0.592335, 1.01725, 2.223404, 2.432124, 1.251632, 1.093335, 0.928403, 0.852925]1104842240.00348658/workspace/gamma_30b_alpha_lora/input/gamma30_v2_4_qwen36_verified_gold_pack_20260615.jsonl50Run strict v2.1 evaluation, then merge/unload or consolidate only if accepted.