Clara (clara-1)

A decision model. You give Clara a record (a support ticket, an agent's conversation, a transaction, an invoice) and the questions your software needs answered. Clara answers every question with a probability for each option you defined, read from the model's label logits. It never writes free text, so an answer is always one of your options and the probabilities always sum to 1.

Question type	You send	You get
noul (yes/no)	the question, optional definitions of yes and no	the probability of yes
choice	named options with short descriptions (2 to 255)	a probability for every option
score	ordered levels, lowest first	a probability for every level and the expected level

Clara is built by Saanora. It is a fine-tune of h2oai/h2o-lightning-4b, which is a fine-tune of Qwen/Qwen3.5-4B. It speaks the open System One wire format (POST /v1/systemone), so clients such as JevBench work with it unchanged. Not affiliated with H2O.ai, the Qwen team or the makers of Jev.

Hosted: the same model runs on the Saanora Platform with API keys, prepaid credit, teams, projects with spend caps and request logs: $0.10 per 1M input tokens, $0 output, and $0.20 of credit when you sign up, no card needed.

Results

Measured by Saanora on public benchmarks the model never trained on. Rows for other models are each benchmark's own published results (October 2026). Two sets of numbers for Clara, because they differ slightly:

• These weights (bf16, as published here), run with the System One server in this folder or in process;
• The hosted API (FP8 on the platform), which is what api.saanora.com serves.

Tool routing, AgentDecisionBench (3,939 agent states: pick the next tool, finish, or ask the user). Official metric: the mean over each task's occurrences, then over tasks.

Model	Accuracy
AgentDecision-4B	99.1%
StartLux-27B	93.9%
OpenJev-27B	92.4%
Clara, these weights (one-way)	92.1%
Clara, hosted API (fast mode)	91.9%
Jev 1.13	91.5%
OpenAI Decisions API	90.9%

Coherence, JevBench-mini (1,200 metamorphic tests; 95% intervals):

Model	Overall
Open-Jev-9B (leaderboard #1)	83.9 [82.5–85.2]
Clara, these weights, two-way	83.6 [81–86]
Clara, hosted API, consistent mode (two-way)	83.0 [80.6–85.5]
Clara, these weights, one-way	81.8 [79–84]
Clara, hosted API, fast mode (one-way)	81.3 [78.7–83.8]
JevAny-Gemma-4B (leaderboard #2)	81.2 [79.0–83.2]

Against its base (same harness for both):

Test	h2o-lightning-4b	Clara
Customer-service yes/no, held out of training (n = 856)	90.4%	94.9%
JevBench-mini, two-way	83.1 [81–85]	83.6 [81–86]
AgentDecisionBench, one-way, scored per distinct question	87.4%	93.6%

Speed: on the hosted API, a record with three questions took 144 ms of model time (fast mode). Format errors: 0%, by construction.

How to run it

The System One server (both reading modes)

This folder carries H2O's serving files unchanged (serve.sh, h2o_lightning_shim.py, pooled_readout.py, box.py, assets.py, serve_config.json) and label rows recorded for these weights (label_rows.lightning). On one GPU with 24 GB or more:

pip install vllm==0.30.0
hf download Saanora/clara-1 --local-dir clara-1
bash clara-1/serve.sh                              # POST http://127.0.0.1:8741/v1/systemone

serve.sh serves two-way reading by default (bidirectional attention over the whole prompt, the more consistent mode, what the hosted API calls consistent). TWO_WAY=0 bash clara-1/serve.sh serves one-way reading (ordinary left-to-right, what the API calls fast). The server checks that label_rows.lightning belongs to the model.safetensors beside it by its SHA-256, which is why the weights ship as one file.

A request:

POST /v1/systemone
{"state": "Ticket 48213: our nightly export has failed three nights in a row. Finance needs the file by 9am.",
 "questions": {
   "failing":  {"type": "noul",   "instructions": "Is a scheduled export currently failing?"},
   "team":     {"type": "choice", "instructions": "Which team should handle this ticket?",
                "criteria": {"engineering": "Broken product functionality", "billing": "Invoices and refunds"}},
   "severity": {"type": "score",  "instructions": "How severe is the business impact?",
                "criteria": ["none", "low", "high", "critical"]}}}

Plain transformers (one-way, one question at a time)

examples/clara_readout.py builds the same prompt as the server and reads the same label tokens, using the server's own functions, so its answers match the server's one-way path:

pip install "transformers>=5.14" torch numpy
python clara-1/examples/clara_readout.py --model clara-1

The core of it:

prompt = render_chat(cfg["prompt"]["system"], user_turn(record, question, names, descs, labels), cfg["prompt"]["prefill"])
logits = model(**tok(prompt, return_tensors="pt", add_special_tokens=False)).logits[0, -1]
logprobs = group_logprobs({i: float(logits[i]) for g in groups for i in g[1]}, groups)   # softmax over the asked labels
answer = format_answer(qtype, names, descs, logprobs, cfg["temperature"][qtype], cfg["noul_floor"])

Two-way reading needs the server's pooling setup and is not shown in plain transformers.

On a laptop

GGUF builds for llama.cpp are in the companion repository clara-1-GGUF: Q8_0 (4.6 GB) and Q4_K_M (2.8 GB), one-way reading only. On 14 test questions, Q8_0 moves probabilities by 0.002 on average and Q4_K_M by 0.013, and neither changes an answer: Saanora/clara-1-GGUF.

Intended use

Routing, triage, moderation, risk flags and agent control: any decision about a record that is yes/no, one of a list, or a point on a scale. Use the probabilities with thresholds you choose on your own data.

Validate before you automate. Benchmarks describe other people's records. Before Clara acts on its own, run it on a few hundred of your own labelled records and pick a threshold for each question. Do not use it as the only decision-maker for decisions with legal or similarly significant effects on people (credit, employment, housing, insurance, healthcare, legal status) without human review.

Limitations

• Asking the user. On AgentDecisionBench states where the right move is to ask the user, Clara is right 59% of the time, below its base (63.5%). It tends to pick a lookup tool when a detail is missing. Add a separate yes/no question ("is a required detail missing?") with a threshold and check it first.
• Score questions (a point on a scale) are the least reliable type. Prefer yes/no or choice where you can; for a scale, sum adjacent levels' probabilities instead of reading the expected level alone.
• Probability coherence (a question and its negation summing to 1) is the weakest JevBench dimension, as for every model on that board.
• Workflow final actions. On our 99-case business-workflow evaluation, per-question accuracy matches the base, but a final action computed from five answers by a code rule is 3 points below the base (the interval includes zero).
• English only. A record longer than max_state_tokens in serve_config.json (32,000 tokens) keeps its head and tail around a "..." line; details in the middle are lost.
• Text only was trained. The vision tower is the base's, unchanged and untested by us.
• One seed pair. The release averages two training seeds; differences under 1–2 points are within noise.

Training

LoRA (rank 16, alpha 32, on the language model's layers only) on h2o-lightning-4b, trained with a soft cross-entropy over the option-label logits. Two seeds were averaged and merged into the weights. Training data, all with recorded provenance:

• ZefanCai/Open-Jev, release-v2-redistributable (CC0-1.0 generated content; the customer-control-v1 and customer-service sources were excluded);
• procedurally generated IT-operations tool-routing states (our code);
• tool-routing plans and multi-question workflow records labelled by a self-hosted Qwen3.5-27B-FP8 (Apache-2.0 weights) on our own GPUs.

No outputs of Claude, Gemini or OpenAI models were used for training. AgentDecisionBench (CC BY 4.0) and JevBench (Apache-2.0) were used for evaluation only.

License

Apache License 2.0, the same license as the base model, for research and commercial use. NOTICE records the attribution to h2oai/h2o-lightning-4b and Qwen/Qwen3.5-4B, the files carried over unchanged from H2O, and Saanora's changes.

If you'd rather not run it yourself, the same model is hosted on the Saanora Platform.

Citation

@misc{saanora2026clara,
  title  = {Clara: a decision model with a probability for every option},
  author = {Saanora},
  year   = {2026},
  url    = {https://api.saanora.com}
}