Sophea-Nemo-3-Nano-v1

ayoubkirouane/Sophea-Nemo-3-Nano-v1

Sophea-Nemo-3-Nano-v1 is a text generation model from ayoubkirouane. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.

<div style="display:flex;align-items:center;gap:18px;padding:16px 20px;margin-bottom:16px;border-radius:14px;background:linear-gradient(100deg,e8f5ec 0%,f7fbf8 62%);border:1px solid c6e7d1"<img alt="KIEFERSA" style="f…

Downloads · 30 days

6% of all-time downloads

All-time downloads

192

Public

Parameters

31.6B

63.2 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors63.2 GB · 100%

At a glance

Task: Text Generation
Library: transformers
License: apache-2.0
Model type: nemotron_h
Access: Public
Created: Sep 16, 2026
Updated: Sep 16, 2026
SHA: d1d51a27

Base models

nvidia/NemotronH-30B-A3B

Model card

From the Hugging Face model README

<div style="display:flex;align-items:center;gap:18px;padding:16px 20px;margin-bottom:16px;border-radius:14px;background:linear-gradient(100deg,#e8f5ec 0%,#f7fbf8 62%);border:1px solid #c6e7d1"><img alt="KIEFERSA" style="flex:0 0 auto;height:58px;width:58px;border-radius:12px;object-fit:contain" src="data:image/webp;base64,UklGRgQNAABXRUJQVlA4WAoAAAAQAAAAxwAAxwAAQUxQSIsKAAABsMf/nyE52q+qZjd2nluEh00OsY2zbdsOH51t27Zt2744F1vrrqrf69X1m+mumq4K/ooISG4bSZKUmIIOeag1w1Ud8wFwPDgH2Om9zYFD2AcXDIa/hdWdgAff+r6oNK7pDDzw1uPxRlQRVoccFlu3e+pj0zrksByD8huqjSFSAm0c2l+8ilrIEQxaT1qC2ljAYYJD89PnxaYQQ05spcdMp5ZEcI0f/CtqGVsygTXY4+vYJGLI4QJg7CeIxkIOFwyGvE0tLeG0Xs8rVLGFHBZb1aORCXAYcniOQdfb61FJjXaEcEb5NeuoWRI+63DhShrgrAmdtZy4iAa4kJPj0PSMOdRCjuCQO3oaDXAhR3Dg+/9Ko0/IERxgl29pDA05QgCM+ohGn7DDYPDr1EJPr2cUiT7hp5aEhvUBZMZ6AY3rC3BjkLVdga0PkFjTGUqCDykOMw9kwEVwKIL9c1zX1d/saIJUsCFvE4sntILyFYjvjgQmeFBwbCsuaAMCyldhhOqVAYERp7bm2s3MKTMoKTF6apugiEOrvaMr+XbADJq8JtXf391IQImt4aEqYpBH3kmtuaXS7E4gkRL1s73pBS8IEY1rLusYCHFi+PoQurwnkjdh+dQ2wHP+i71p/GgM/XaQggITFpzdIpaQoWL7ZlcaGtKD5LPnrBOaeB8tbO3XAzg5LylIsYe/HyY8FwuLNE4/uoSck5SkkO/38jsk2djc05uSK5uSlOvQZzsAcBESdGyLJrckP4hZkFLeGw7AuZ+kDnAXt6fVyIa0ol7u62u0SGfrri2nM5xDQpJ8oqefksbq7+hOk4Eddgmw4e7uwHLMa6TC6NEe9DLaYyfV13sY95JMv5AX4FxgK6sv7ZiFCVwkDFdIifj2EBp93GAvS6a29ivuFbRPx1Jzi135nndWsyKXMdZvxLDCo7kDlNT43Z5AjjYTFDiW6ceVFFEYNJmOqtCthdgLuCXSBLiDaIBzi5sJv5PjKRalf6PU+UNptZ0lKrZpx5WSo8wA6eLeHhBLkZiWsNGWG/P8nEmfnwySd499Og5AsMyDSye3ISuzexyKQnx7KGSfSzvS6FMEHIt+ui2wjEPrffFxUcaGg8g09IHJPqg1DgGe9Q34AeLQTSF0Wr5egOfMJlKBh5S4QQ+f9zNqqfwkS9bzMV3TCU4yHXsq1LDYut9Ti7i6G0Dzc0gHpX9kxcpvXIeqEas7QQmDNv9bhipSgYUE2PaXrUIVadJHwUhv7pWrUUU6oJAvHi0nLzUnK78fhJ7STrdWo4rCSWzNzphvFpSkvhYuGGxxv+ldDyM5DqXHzEAtdZr+HBZLr6ckShk+BAdx0O+mTCX1GSUUslcwFk8oprE9v0EtpU2/lBAAI99DLWW44AJg/EfmzKTr+0qSnb4y0SJMCMFg2FsapXTRvyY48L1/MhPCA4+t9wsKpXTVhyc4iCOmmZAUFlhsVY81oops+glTSZNTZpv1NBzw2LreVYdKard9kSyWFhMWmLgXAmg+KLtuHSpZhP5OJhi0vWC5CUn+w3Ic2l9IAhzaYXfyr1pjdttzBIfWE5bQ/OwAiwldbiMPQyZxGeBO+we1MQdYylYPNpoJGcSZ5Y6ZRs0BDhao7Z5VqKSPCA7soLz3fUc4kIFvIErpG5wD7JYX4LIB/WFw9IdGMkRfKM1ZjPxgNuZj+tOCS9xck12/NtckM6T+HSwh8r9F7z23uHpGDjTPiMwCiPjQ1TdcbzGM9XqOrgWOcSe5o/5CFRURdyNvbZYa3eO0hsxGHJIFZGQ1utxeRz+2uMettJywbCSIjG9WUssIViGpvD2wjKONFQfXAgCQdRB9AICxzIO+ALAhy7ouwNYHSB1vSsKKRSvuzsAEDx8OOgWnbr38jQFG/CC7tvaazaB8BTY+uV0RGkkg43nIsdXe0RkEVKzECOvv28JMCMbGsTU8vCUwIaB8FfkAUn1LJbAc84mbJ/33PxbD2eIoUT5rbmGe10dBH9ErOwITzBM02vYTOjKF+PpQc/sW7AchE5b9p7VLyfj7tKMZ+CFpxU3ua9GRxoVnN3fXHpzx7xsuTOPXuwFpnEyETNA4+/gmwEXoULH9sj8n5yeBhD3745CSWHzHQQvzkSXAcwCJJO7dD3uzWEKFMq24pzUl5yUVifLpjgBChAgd25KJLcnVTUWqNejdkaYgBwZSlVZc0o7U15SkrGwv9TfiK+ltzfVltCKlJW11i57sacRPUge4O7tScwpipLD+/m5mh4OAVBg90oNeTlvSnvrqGyqM+I6WCvWzfelN6Yo0O77mso4muHqNlIhvDqXRxwU2E5ZNbm1CkrdIqfGTsdTcYFcq5p3ZHLjwExXbd7vRIuoK29I9/YQmsfiHijT+diAnR+wKF/LrISIWbyh03Y7LkaMtOkn31V4mJHlC3oow70wa4IpOitXq0/EAnHsBscVTWpF1zS3u5O2hRjwg0rjq4va0YjrGoegX+5jDyv6m+vpKenjOcSgK5WNVkH3u7kZrvnPcf6q6vSWwjEPLbVFwLsNAZAJbyz6oNQ4BnvmNHyAO3RRCp+XrBXgOylZiowo8pFBvddMlc1BL7SdZsorra+sqm/5vuZFQw3IcOly2EnFNV4COl5Om2gBDvni0nLIEVSNWd4ISBpW31hjJNn+j1PlDaRWTDWt21jyTbEgfBeOCQff761FGOqib2EqOm2XyWaF+ECPbPtWISnpF8Y0f+DtqqQr1tRRoqn1Vo5TucC7de/YoPEozYHt8h1qqpP4cKgCj3sVYHOH14AJg3MeopUzbZyQEwM5foJYqk/CkUWwb/qa5zjb9UoID3+dH1FKGa8MFg74vmOfOtu9LcMgd9reZECZYbD0fbUQVpe1fS5QmJ80166gd/lrXu2tNhU3Zh5cqYzQ/b6GRoEC+eJRdvxaV1O76CZlg0OaC5SZaeEyqi9ruEtJQbNUXmVI6Xk3iXiAQDFpNWYramAUWt08nGvcCQDyj2anz6NuANaknVD3QgCryHcGh9OgZtJxaYlkkej2tUEmfEQLEAb/QUOAASxn0GqKUviI4wO7f0EBQLAqHpFEfGfERYQJc3jEXHwATwnb+2lwL3+CCweA3dGwai0tS3Nv/F1OQC+CJ9Xpe0hcYd7gRceTfRnyBxbblIzTAOcRd7Why2hwzgeDBjM531JHo4xSXnxNaTFiEKvIABmXXVNNPR25xm63aX7AccUjmaX/BSmqucZ51r6kbASLjzKMBrhi4lqoyYBkHpUbMOgCMzMg4CtEHADjLPOgLABuyrOu0fkDpmi7rAUi2WfgvYIGHvNnV3do5SH/gfEbjw1saCxGOZ6hntzWZM+xIifqNwdTChEPDD8cA8NgCDnnL/nJXaiFHxfbL/oy0f4YMNzbtqFJiIUdFCv85tTmxgEPeTxdPaEUs5OhI4coL2pJ/WAghdrbu2jLyNSqMw8Zq7+pCLeSYAPdYFbWQIyXK5/rQ6BNypER8fQi1oJLKPhoPYCzkKKnxm92AFKrgkmi/HSiIBZgEm3F0KbGQoyKN805tSizk6EjhkqktSfQJNvkB7pJ2wIyFG2JrbyynoSHkmC9hd3WjFnYweqQnDQ2h59W+1IIPAABWUDggUgIAAHAdAJ0BKsgAyAA+MRiJRCIhiBJcEAGCWlu4MAAZsUGEUQU6Vr/HVACel73mNKbZii7/0o6X9Y+s6FoAm3srqdXIcWd/45kb7mdOviNuuq60W3qXx/nI9sR42ICkrOwkSyYty0cun8cMpqXF0fh8fW3dSwaV0akfevRFNfnTLgy9bQM6IEWlj9O95hHRhKjNbDCS4qWJnQbJto47spFiqntPEoRO945gp7f+FzLmeAJQakEjD+xgQDxMOYrHw79WEkLQkKplqTo/uTX4n54eEP0N4epeSNwNE7Cl7q0Xi32YrSRma3RieLfasZELkHnrVj60WkAA/uouoblbKpSNm/wu8g0tISmLMdPrSIKVi8oMaNvyXkZc6zskreB0cAODf99ubzUwj0N8IJywGgCWpByLV1Kx+G8TMEVJa0HKGlyheflfEABBIEOGipfrRMJEDIiV7tNyTCRkHRlacBBPJNrPF6gu5KqHk0mGh2xBZYDKEWXI1A8R9WyRig0s6CIELwgJcqvHXXixg1yTYYd3Ja2rm0FNGHOb8MdNErR6nH3DYmradAb7JnP/RNqQ3gvhl/pA7ygUm6CIWAO/JF0d93ZViT6KqEgvSVdzZIwTzPR1TQ5YDMPK1u42qqN2SqTRwtzHrtold7EeIrx7plafSjlNtoSSwuC99MOigEHtk+7uPfbMdQA6uO4+wAsQ6bfVGy+dpyM63pbfBdYa2IbgqA8ZA3EUsyDx6GwlVker0XVZhE/HSCyOiMAVqYYDYXpIECKSg6k4BD3OKMJ7nJlAAAAAAA=="/><div><div style="font-size:1.24rem;font-weight:700;color:#003713;line-height:1.2">Sophea-Nemo-3-Nano-v1</div><div style="font-size:.86rem;color:#3a5a47;margin-top:2px">Greek and English reasoning fine-tune of NemotronH-30B-A3B: Mamba/MoE hybrid</div></div></div>

📄 Released with the paper Thinking in a Low-Resource Language: What SFT Builds, What RL Fixes, What Accuracy Cannot See (arXiv:2608.17744). Every number on this card is measured, contextualized, and caveated in the paper; read it for the full evaluation protocol.

Creator: Kiefer SA (Sophea AI Lab, Athens)
Base model: nvidia/NemotronH-30B-A3B (Mamba/MoE hybrid, 31.6B total / 3.58B active, 6-of-128 routing)
Languages: Greek and English (the reasoning trace follows the question's language)
Decoding: thinking: the reasoning trace is carried in <think> blocks and follows the question's language; direct mode is clean (zero unrequested traces across the study's <think>-carrier direct-mode generations, n=9,751)

Serve thinking, and mind one environment note. Under the library versions the study used (transformers 5.12.1), the shipped remote modeling code trains correctly but fails at generation (prepare_inputs_for_generation indexes a cache_position that generate() passes as None); the study evaluated with trust_remote_code=False. mamba-ssm is optional: installing it changed step time by 0%.

Intended use

Greek and English reasoning where answer-channel integrity matters most: 0.0% leak, and negligible fallback on math and commonsense (1.7 / 1.8%)
Deployments that also need the base's general Greek ability: the fine-tune gains +3.8 Greek macro on the Titan-1 suite (largely catch-up against a low base)

It does not obey an instruction to reason in English on a Greek question (0.0% override; the reverse direction works at 87.5%). Do not deploy it where an operator must force English traces. Not evaluated for safety-critical, legal, or medical use.

Fine-tuned from NemotronH-30B-A3B; shipped merged to dense weights.

Evaluation

Arrows mark the desirable direction for every metric: ↑ higher is better, ↓ lower is better.

Headline scorecard

The claim replicates on a third, non-transformer architecture: the base reasons in Greek on 0% of the benchmark, the fine-tune on 97.4%, at an accuracy cost of −0.6 pp on that instrument, with generation-cap truncation falling 10.8% → 2.6% and measured loops in its traces dropping 15.0 → 0.0 per thousand rows.

Per-benchmark detail

Per-domain, Greek think lane

Read the logic cell carefully: the 9.8 macro-recall is substantially a format floor, not a capability reading. The release's fallback is negligible on math and commonsense (1.7 / 1.8%) and 39.9% on logic, the one domain where its answer form breaks; logic is also the hardest axis for every arm in the study, base models included. On the override probe this family returns 5–9% of rows too short to score (Qwen: 0%), consistent with a low reasoning ceiling.

Forgetting check: Titan-1 suite (non-reasoning mode)

This suite exists to answer one question: did the Greek fine-tune damage general ability? It did not. Against its own base, across 14 bilingual NLU benchmarks: +3.8 Greek / −2.0 English; the fine-tune gains on the Greek side and the English movement is within the range seed noise alone produces on this class of suite. Read the numbers below as a forgetting check, not a capability leaderboard.

<div style="overflow-x:auto"> <table style="width:100%;border-collapse:collapse;font-size:13px;font-variant-numeric:tabular-nums;font-family:-apple-system,BlinkMacSystemFont,'Segoe UI',Roboto,sans-serif"> <thead><tr> <th style="padding:10px 8px;text-align:left;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">Benchmark (Titan-1 suite)</th> <th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">Base</th> <th style="padding:10px 8px;text-align:center;font-weight:500;border-bottom:2px solid #0a8043;color:#0a8043;font-size:14px;">Sophea-Nemo-3-Nano-v1</th> </tr></thead><tbody> <tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">ARC-Challenge-el</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">54.2</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">59.0</td></tr> <tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">ARC-Easy-el</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">62.3</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">70.1</td></tr> <tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Belebele-el</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">67.0</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">78.0</td></tr> <tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">GreekMMLU</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">58.8</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">53.3</td></tr> <tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">HellaSwag-el</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">39.2</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">47.6</td></tr> <tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Medical-MCQA-el</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">27.8</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">31.2</td></tr> <tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">TruthfulQA-el</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">31.7</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">33.2</td></tr> <tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Winogrande-el</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">53.3</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">58.3</td></tr> <tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">MMLU-el</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">43.2</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">40.8</td></tr> <tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);"><em>Greek macro</em></td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);"><em>48.6</em></td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;"><em>52.4</em></td></tr> <tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">ARC-Challenge (en)</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">88.3</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">82.2</td></tr> <tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">ARC-Easy (en)</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">96.2</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">93.4</td></tr> <tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">HellaSwag (en)</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">67.9</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">73.1</td></tr> <tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">Winogrande (en)</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">69.9</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">72.9</td></tr> <tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);">MMLU (en)</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);">69.9</td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;">60.5</td></tr> <tr><td style="padding:7px 8px;padding-left:20px;text-align:left;border-bottom:1px solid rgba(128,128,128,.15);"><em>English macro</em></td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);"><em>78.4</em></td><td style="padding:7px 8px;text-align:center;border-bottom:1px solid rgba(128,128,128,.15);font-weight:700;"><em>76.4</em></td></tr> </tbody></table></div>

Register control improves 40 → 44 of 46 and grammaticality roughly doubles, 13 → 27 of 58, driven by the agreement axis (judge-scored counts, directional evidence only).

Usage

Serve with vLLM (OpenAI-compatible):

vllm serve KIEFERSA/Sophea-Nemo-3-Nano-v1 --served-model-name sophea-nemo-3-nano-v1

Client (OpenAI SDK):

from openai import OpenAI

client = OpenAI(base_url="http://localhost:8000/v1", api_key="EMPTY")
resp = client.chat.completions.create(
    model="sophea-nemo-3-nano-v1",
    messages=[{"role": "user", "content": "Ένα κατάστημα πουλά στυλό προς 1,50 € το ένα. Πόσο κοστίζουν 12 στυλό;"}],
    temperature=0.7,
)
print(resp.choices[0].message.content)  # <think> trace in the question's language, then the answer

Steering the trace language (partial, measured per checkpoint): append "Think step by step IN ENGLISH." to a Greek question, or "Σκέψου βήμα προς βήμα ΣΤΑ ΕΛΛΗΝΙΚΑ." to an English one. On this checkpoint the English suffix is not obeyed (0.0%); the Greek suffix on an English question is obeyed on 87.5% of items.

License

The fine-tuning additions are released under Apache-2.0. The model inherits the base model's license terms (nvidia/NemotronH-30B-A3B); verify the base model's license before use.

Citation

This model is a release artifact of arXiv:2608.17744. If you use it, please cite:

@misc{kirouane2026thinking,
      title={Thinking in a Low-Resource Language: What SFT Builds, What RL Fixes, What Accuracy Cannot See},
      author={Ayoub Kirouane and Christos Petrocheilos},
      year={2026},
      eprint={2608.17744},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2608.17744},
}

Task: Text Generation
Library: transformers
Type: nemotron_h
License: apache-2.0
Languages: el, en
Created: Sep 16, 2026
Updated: Sep 16, 2026