GoMini × AgentGo
Local choices. Independent checks. A larger model when needed.
322M parameters · Full fine-tune · Native + Q8_0

How it works · Results · Download · Quickstart · Limitations

GoMini v4

A small, local decision model for a bounded browser loop.

GoMini is a full fine-tune of THX-01 for selecting the next action from an explicit set of valid candidates. It is designed for AgentGo browser workflows: a larger model supplies a bounded subgoal; GoMini chooses among prepared actions; AgentGo executes permitted actions and checks their effects. Ordinary verified steps can stay local, with the larger model called when the subgoal finishes or needs help.

Created by Mhmadhassnz.

Development release, not a passed production gate. V4 was selected using development data. Geometry selection reached 85% on 20 local episodes; adapted human website decisions reached 82.77% on accepted inputs, or 72.42% including packing rejections. Independent final validation, calibrated fallback reliability and actual Q8 browser accuracy remain pending.

Model	GoMini v4
Foundation	doofz/THX-01 · ModernBERT decision architecture
Parameters	321.9M total; 321.7M optimized
Fine-tuning	Full encoder, embeddings and decision head; no LoRA
Validated task	One bounded choice question named next
Input contract	1,024 packed tokens; 2–24 explicit choices
Formats	Complete native Safetensors checkpoint + Q8_0 GGUF
Training hardware	AMD Radeon RX 7800 XT · 16 GB
License	Apache 2.0; dataset attribution retained separately

What GoMini does

GoMini scores supplied candidates in a non-autoregressive forward pass. Choices can represent a click, a supplied field/value pair, a prepared read method, or explicit wait, done, escalate and stop controls. It does not generate an arbitrary plan, new selector, tool argument, value or chat response.

The release contains reference selection clients, not a complete production AgentGo integration or permission system. Every client response carries execution_authorized: false; the supervisor remains responsible for execution.

Results at a glance

These are native v4 development results. Local suites used actual model selections and AgentGo execution/readback on owned procedural pages. They are not public-web accuracy, independent test scores or Q8 browser scores.

Development evaluation	Result	Scope
Single-action episodes	400 / 400 · 100%	Local click/fill/select/check and control scenarios
Multi-step form episodes	133 / 133 · 100%	471 decisions; local draft and completion checks
Reveal-only viewport episodes	200 / 200 · 100%	352 decisions; reveal without activation
Strict-navigation episodes	300 / 300 · 100%	650 decisions; prescribed scroll/back/wait policy
Prepared read-method episodes	377 / 380 · 99.21%	19 canonical read methods, prepared arguments
↳ Element geometry subgroup	17 / 20 · 85%	Confusion with viewport/HTML reads
Mixed offline accepted decisions	17,502 / 17,590 · 99.50%	Mostly procedural/synthetic examples
↳ Adapted human snapshots, accepted	365 / 441 · 82.77%	Mind2Web-derived, target included in shortlist
↳ Adapted human snapshots, all raw inputs	365 / 504 · 72.42%	63 packing rejections counted as failures

Zero false-done predictions were observed in the listed native action, form, viewport and navigation suites. These include scenarios where a correct stop or escalation is success; they do not prove universal task completion or reliable failure detection. Navigation's oracle requires explicit scrolling even when a browser click could auto-reveal its target.

Q8_0 delivery smoke: 64 attempted synthetic/TRAIN-derived inputs; 56 supported inputs matched both exact token IDs and native top choices. Eight inputs were rejected before Q8 decision inference by the packing guard. Maximum absolute probability drift on compared inputs was 0.0008011311. This checks conversion and interfaces, not browser success.

Read EVALUATION.md for every native episode subgroup, all 19 read methods, offline source breakdowns, rejection counts, methodology and remaining evaluation. Machine-readable summaries and the original packing failure are in reports/.

Download

Artifact	Size	Use
native/model.safetensors	643,835,524 bytes	Full native checkpoint; download its configs/tokenizer/runtime too
gguf/GoMini-v4-Q8_0.gguf	356,866,560 bytes	Decision-capable GGUF runtime only
adapter/	Python helpers	Shared packing and resident selection clients; no learned adapters
runtime/thx01/	Python implementation	Bundled upstream native inference code

Use revision v4 for the tagged release. Checksums are in SHA256SUMS.txt, with conversion identities and status in PROVENANCE.json. No Q4 or intermediate F16 GGUF is distributed.

Quickstart

Use a compatible GPU PyTorch environment with Transformers, Safetensors and NumPy. The recorded environment and Q8 server requirements are in ENVIRONMENT.md. CPU portability and other GPU stacks have not been validated by this release. It is a custom decision checkpoint; loading it as AutoModelForCausalLM or sending chat-completion messages is not supported.

Download and run the included single-request example with that environment:

import subprocess
import sys
from pathlib import Path
from huggingface_hub import snapshot_download

root = Path(snapshot_download("Mhmadhassnz/GoMini", revision="v4"))
subprocess.run([sys.executable, str(root / "examples" / "choose.py")], check=True)

The example reads example_request.json, serializes it to one line and asks the native chooser to select a candidate. It does not execute browser actions. For an integration, keep adapter/choose_native.py resident: write one compact JSON object per line to stdin and read one JSON result per line from stdout. Diagnostic messages use stderr.

For Q8_0, first start a compatible ModernBERT decision-capable llama.cpp server with /tokenize and /v1/systemone, following ENVIRONMENT.md. Its binary is not included. Then run:

subprocess.run([
    sys.executable, str(root / "examples" / "choose.py"),
    "--q8", "--url", "http://127.0.0.1:8765",
], check=True)

Use adapter/q8_client.py for a resident guarded Q8 client. Keep its packing and runtime token checks: some literal-special-token/trailing-whitespace inputs do not match native per-piece packing and must escalate. Native/Q8 agreement on the limited smoke suite does not establish equivalence for every possible input.

Integrating with AgentGo

The recommended loop and implementation details are in LIMITATIONS_AND_AGENTGO_INTEGRATION.md.

1. Have the large model specify the subgoal, completion predicate, permitted tab and origin, exact supplied values, effect permissions, deadline and step budget.
2. Build a fresh compact observation and bind valid candidate keys to prepared browser operations. Include explicit wait, stop, done and escalation controls.
3. Use the shared gomini.agentgo/1 packer. Keep the 512-token instruction/options head budget, 1,024-token total, maximum 24 choices and 48 tokens per rendered option. Reject oversize input rather than silently truncating it.
4. Independently check permissions, tab/origin/revision, candidate membership, exact field/value binding and attempt limits before executing any selection.
5. Verify effects from browser evidence. Stop on cancellation; return compact evidence to the larger model on verified completion, ambiguity, missing information, repeated failure or exhausted budget.

For a known fixed request such as an element rectangle, deterministic dispatch to the correct prepared method can be better than asking a model to rediscover the mapping. A successful command or HTTP 200 does not prove goal completion.

Limitations

• General browsing is unproven. Local fixtures share families and some markup; split-separated local labels are not independent real websites or accounts.
• Human transfer is weaker than the procedural average. The adapted dataset includes the annotated target in each shortlist. Its scores are not official Mind2Web benchmark results or full browser trajectories.
• 1,088 registry names are not 1,088 verified tools. Training includes synthetic description routing, including aliases; actual read-method checks cover 19 canonical methods, with prepared arguments and no write-operation coverage.
• Confidence is uncalibrated. V4 uses temperature 1.0; confident mistakes occurred. Probability, margin and entropy must not grant permission. The unused auxiliary action head was not supervised and must not gate safety.
• Fallback is an unmeasured property. Do not assume every failed action will request help. Validate independent checks and larger-model handoffs separately.
• Q8 browser evaluation and final audit remain pending. Packing guards reject some inputs; reference clients are not a hardened supervisor. No general >90% success, latency, multilingual retention or production-readiness claim is made.

Training and provenance

V4 continues the prior native checkpoint with 53,146 accepted unique training records, combining procedural examples, synthetic registry routing, adapted Mind2Web snapshots and verified decisions on owned local AgentGo fixtures. It reached epoch 2 / step 5,660 using AdamW, BF16 autocast, FP32 optimization and a stable log-score plus spherical-score objective. Embeddings and all encoder blocks were trainable; this was full fine-tuning.

See TRAINING.md for the recipe and source counts, and native/training-settings.json for exact data file hashes and settings. No raw website corpus or complete collection harness is hosted here. Development influenced selection; frozen calibration/final-test records were not consumed for this handoff. Training is paused at v4.

Credits and license

GoMini builds on THX-01 by HAL-X AI, with its mmBERT encoder attribution preserved in NOTICE. Upstream weights and bundled runtime retain their Apache 2.0 license. The pinned foundation revision is b323afcef149f9288a9d0daef4badd67b9b87c61.

Mind2Web is credited to Xiang Deng, Yu Gu, Boyuan Zheng, Shijie Chen, Samuel Stevens, Boshi Wang, Huan Sun and Yu Su (2023). Its dataset is CC BY 4.0; the local adaptation changes representation, packing and split assignment. See THIRD_PARTY_DATA.md and the official Mind2Web project.

AgentGo is the browser integration target and execution environment for the reported local fixture checks. AgentGo browser software is not distributed in this model repository. The original AgentGo logo and ember-on-black palette are used for the visual identity of this card; see visual asset credits for provenance, the separate logo license and illustration disclosures.