This is sweep arm natural_tool2x from a four-arm controlled SFT study for Android GUI use. The language model was fully fine-tuned while the complete Qwen3-VL visual tower, merger, and deep-stack mergers remained frozen. The uploaded weights were verified byte-for-byte against the pinned base for every model.visual* tensor, while at least one language-model tensor was verified to have changed.

Training contract

Base: Qwen/Qwen3-VL-8B-Instruct at commit 0c351dd01ed87e9c1b53cbc748cba10e6187ff3b
Dataset: luca0621/appgen-sft-ngc-v1, config ngc-f-qwen3-normalized, revision 769ea99dbc4ff190048ae0db37eb6310dba595e0
Dataset variant: natural
Dataset SHA-256: 8890e2fe596e4d78714ce4309f2a59ed9632dd5d656f758c8dc4608d46a9c6bc
Exposures: 3,588; unique semantic examples: 3,168
Direct-grounding exposures: 400
Completion-retention replay exposures: 420
Coordinates: normalized 0–1000
Prompt variant: proven; SHA-256 67ff8adb0e78a617f3d0edcf196d4e4cc3239967a8c30619fc5afc14484ee8c0
Optimizer: full-language AdamW, learning rate 5e-7, cosine schedule, 5% warmup
Batch: microbatch 4 × two GPUs × gradient accumulation 4 = global batch 32
Epochs: 1; optimizer updates: 113
Frozen: visual encoder and aligner; trainable: language model and LM head
Swift's hermes loss-scale preset was used (the sweep's tool2x arm).

Only the final checkpoint (checkpoint-113) is published in this repository. The exact training system prompt is included as appgen_system_prompt.txt; run_manifest.json records hashes and verification evidence. Intermediate 25-step checkpoints remain local for controlled eval.

Data provenance and limitations

The images are synthetic AppGen HTML-to-PNG screenshots from 50 generated training environments. They are not real-user or real-device captures. The pinned dataset declares no AndroidWorld, held-out, or public grounding benchmark images. No AndroidWorld or real-world benchmark score is claimed in this card; evaluate all four arms under the same decoding and benchmark setup.

This model is for Android visual-agent research, not safety-critical autonomous deployment.

namhokaist/appgen-qwen3-vl-8b-sft-ngc-v1-frozen-natural-tool2x-proven-lr5e7-1ep

AppGen Qwen3-VL frozen SFT — arm B

Training contract

Data provenance and limitations

namhokaist/
appgen-qwen3-vl-8b-sft-ngc-v1-frozen-natural-tool2x-proven-lr5e7-1ep