Saina Helm 3

Helm 3, natively multimodal. Saina Helm 3 reads text, images and audio together in one pass and answers typed questions about them: yes/no, single choice, multiple choice or rating. It returns a probability for every answer instead of generating text, so every decision comes with a confidence you can act on.

What it does

One pass, many questions. Send the state, your images and audio clips, and any number of questions. Every question is answered from a single pass over the shared input.
Reads everything at once. Text, screenshots, photos, charts, receipts and short audio clips arrive in the same request, and every question sees all of it.
Knows when to abstain. In decision mode, low-confidence answers come back empty with a reason, so your system can escalate instead of guessing.
Small and fast. 0.8B parameters; runs on a single GPU, on Apple silicon, or on CPU.

Benchmarks

Saina Helm 3 scores 45.89 on the Decision Index, a benchmark of decision tasks (public index, full 140,620-request suite). Scores are out of 100: each benchmark is chance-corrected, so 0 means random guessing and 100 means perfect.

Model	Decision Index (text)	Decision Index Vision
Saina Helm 3	45.89	47.36
Saina Helm 2 Omni	37.44	46.90

Full runs, self-run with the published harness (run-saina/saina-decision-index).

Intended use

Decisions where a wrong answer is recoverable or reviewed before it takes effect: routing a support message with its screenshot, checking a receipt against a claim, flagging a call whose voice may be synthetic, tagging sounds and images. Validate on your own labelled data, for your own label space, before use.

Out of scope

Generating text of any kind; the model only scores the options you supply.
Final, unreviewed decisions with material consequences for a person (credit, employment, medical, legal).
Identifying or classifying people, or profiling individuals.
Safety filtering or content moderation.
Video, and audio longer than 10 seconds.
Decisions taken directly on untrusted content: text, images and audio can carry instructions that steer the answers.

Usage

pip install "saina[local,media]>=0.9.1"

from saina.transformers import HelmModel

helm = HelmModel.from_pretrained('run-saina/saina-helm-3-0.8b', device='cuda:0')

answers = helm.ask(
    model='saina-helm-3-0.8b',
    state={'claimed_usd': 244.80, 'policy': 'Client meals up to $300 with an itemized receipt.'},
    images=['receipt.png'],
    questions={
        'total_matches': {'type': 'yes_no', 'question': 'Does the receipt total match the claimed amount?'},
        'action': {'type': 'single_choice', 'question': 'How should this claim be handled?',
                   'options': {'approve': 'Approve', 'review': 'Send to a reviewer', 'reject': 'Reject'}},
    },
)

Use the saina package: it builds the exact prompt the model expects and applies its answer head. This model needs saina 0.9 or newer (older versions refuse it); use 0.9.1 or newer, which uses less memory on long inputs.

Where to run it

Runtime	Where
Hosted API	saina.run, model `saina-helm-3-0.8b`
PyTorch / `saina`	this repository
MLX (Apple silicon)	run-saina/saina-helm-3-0.8b-MLX
GGUF (llama.cpp)	run-saina/saina-helm-3-0.8b-GGUF
Ollama	run-saina/saina-helm-3-0.8b, with the Saina Ollama build (branch)
Self-hosted server	`pip install "saina[local,server,media]>=0.9.1"`, see run-saina/saina

Inputs

Input	Formats	Limits
Text	`state` (string or JSON) and the questions	up to the context window
Images	PNG, JPEG, WebP	8 per request, 10 MB each
Audio	WAV, FLAC, MP3, Ogg, WebM	4 per request, 10 MB and 10 s each

Question types

Type	You provide	You get
`yes_no`	`question`	`yes`, `no`, `confidence`
`single_choice`	`question`, `options` (2 to 255)	`probabilities`, `selection`, `confidence`
`multi_choice`	`question`, `options` (2 to 255)	independent `memberships` per option
`rating`	`question`, ordered `levels` (2 to 10)	`probabilities`, `expected_level`, `confidence`