whatmodel.sh / Hivemind 32B Preview

Hivemind 32B Preview

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Hivemind 32B Preview is an open-weights language model from PrentisAI (HiveLabsAI), released April 2026. Its published size is 32.76B parameters. Licence: other (proprietary).

Catalogue identity notes

Maker repo PrentisAI/hivemind-32b-preview @78df78b (requested as HiveLabsAI/hivemind-32b-preview, which the HF API redirects to PrentisAI/hivemind-32b-preview): licence tag other (license_name proprietary), card base_model Qwen/Qwen3-32B, tags emotional-intelligence / conversational. Released 2026-04: month the maker's Hugging Face repo was created (createdAt 2026-04-03). Held at the 2026-10-10 merge because its only result is one EQ-Bench 3 Elo (placed far above its size class); imported under the 2026-10-11 owner decision.

73.9±4.1#51 of 949 LLMsfrom 1 benchmark result across 1 source* thinly tested, treat the rank as provisional

$ how it compares

The LLMs ranked either side of Hivemind 32B Preview, out of 1132 with a score.

#modelscorewhereRAMreleased
53GLM-5.3-Flash Z.ai74±1api $0.24/M +local~103GB2026-08
54Gemini 3.5 Flash Google74±1api $3.38/M2026-05
55Hivemind 32B Preview PrentisAI (HiveLabsAI)74±4local~70GB2026-04
56GPT-5.2 OpenAI74±1api $4.81/M2025-12
57Hunyuan Hy3 (295B-A21B) Tencent74±1api $0.23/M +local~96GB2026-07

see all LLMs · find the ones that fit your machine

$ what it's good at

other tests1 not in a task · 92% of its score
testresultvs the others in its rundifficultycountswhy
EQ-Bench 3 (emotional intelligence, judge Elo)1351.0 ✎▁·▂▁▄▅▅▃█▄▁▁ beats 59/62avg 1044▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 92%typical

$ where to run it

Other ways to run it

  1. Run it as published · needs ~65.52 GB · from huggingface.co/PrentisAI (official)
    uv run --with torch --with 'transformers<5' --with accelerate python -c "from transformers import pipeline; print(pipeline('text-generation', model='PrentisAI/hivemind-32b-preview', torch_dtype='auto', device_map='auto')('Once upon a time', max_new_tokens=40)[0]['generated_text'])"

Official weights and documentation: huggingface.co/PrentisAI ↗

$ how the score was worked out · EQ-Bench 3 canonical leaderbo… 92%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
EQ-Bench 3 canonical leaderboard results1▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 92%MIT
prior · pull toward the middle▇▇ 8%
weighted average= 73.9

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json