New
Everything new in AI as it appears: launches announced by the labs, brand-new models, datasets, papers and open-source tools. Checked every minute or two. Launches feed · Trending · News
Live now
- Connecting…
Just launched
Yesterday
- Model
- Datasetclm-project/jev-boost-kn4-aces-salient-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn4-aces-salient: Qwen3.6-35B-A3B-FP8 state/action embeddings nikitam/ACES (revision b497a6456957a5660ac20b8cac5b5222eb9b669c), split train, config ACES, conve
- Datasetclm-project/jev-boost-kn4-preco-wino-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn4-preco-wino: Qwen3.6-35B-A3B-FP8 state/action embeddings coref-data/preco_raw (revision 02c8ffedb9d0b49168809a572ff553c2f6f9f7ca), split train, converted in
- Datasetclm-project/jev-boost-kn4-synthetic1-crux-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn4-synthetic1-crux: Qwen3.6-35B-A3B-FP8 state/action embeddings PrimeIntellect/SYNTHETIC-1 (revision f08fe8ccdb5be7ea7d3645179aa0361a8ad0a37e), split train, c
- Datasetclm-project/jev-boost-kn4-tabmwp-gsm-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn4-tabmwp-gsm: Qwen3.6-35B-A3B-FP8 state/action embeddings theblackcat102/tabmwp-clean (revision 01b2313e62aa4c62c5a3370a460f04098d4b3172), split train, conve
- Model
- Datasetclm-project/jev-boost-kn4-calcx-gsm-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn4-calcx-gsm: Qwen3.6-35B-A3B-FP8 state/action embeddings MU-NLPC/Calc-X (revision f4f54de8cfd3f6349963e174b5997f4aeb061fc0), split train, converted into deci
- Datasetclm-project/jev-boost-kn4-gsm-variants-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn4-gsm-variants: Qwen3.6-35B-A3B-FP8 state/action embeddings Emulated-Inc/grade-school-math-training-pool (revision 30102ba31e289cb689d17577598520d07700114d),
- Datasetclm-project/jev-boost-kn4-lichess-quiet-chess-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn4-lichess-quiet-chess: Qwen3.6-35B-A3B-FP8 state/action embeddings Lichess/chess-position-evaluations (revision 8c5285259fdbf3218bf2582de9775dd1378f520b), sp
- Datasetclm-project/jev-boost-kn4-lichess-multipv-chess-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn4-lichess-multipv-chess: Qwen3.6-35B-A3B-FP8 state/action embeddings Lichess/chess-position-evaluations (revision 8c5285259fdbf3218bf2582de9775dd1378f520b),
- Model
- Datasetclm-project/jev-boost-kn4-lichess-eval-chess-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn4-lichess-eval-chess: Qwen3.6-35B-A3B-FP8 state/action embeddings Lichess/chess-position-evaluations (revision 8c5285259fdbf3218bf2582de9775dd1378f520b), spl
- Datasetclm-project/jev-boost-kn4-allocation-musr-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn4-allocation-musr: Qwen3.6-35B-A3B-FP8 state/action embeddings Emulated-Inc/narrative-reasoning-training-pool (revision bce45498fc4ba8a281e23839a455029c565a0
- Datasetclm-project/jev-boost-kn4-placement-musr-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn4-placement-musr: Qwen3.6-35B-A3B-FP8 state/action embeddings Emulated-Inc/narrative-reasoning-training-pool (revision bce45498fc4ba8a281e23839a455029c565a04
- Datasetclm-project/jev-boost-kn3-megascience-gpqa-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn3-megascience-gpqa: Qwen3.6-35B-A3B-FP8 state/action embeddings MegaScience/MegaScience (revision 8df5586005374acba25aecc4f5469ce30fec605c), split train, con
- Datasetclm-project/jev-boost-kn3-megascience-mmlupro-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn3-megascience-mmlupro: Qwen3.6-35B-A3B-FP8 state/action embeddings MegaScience/MegaScience (revision 8df5586005374acba25aecc4f5469ce30fec605c), split train,
- Model
- Datasetclm-project/jev-boost-kn3-webinstruct-gpqa-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn3-webinstruct-gpqa: Qwen3.6-35B-A3B-FP8 state/action embeddings TIGER-Lab/WebInstruct-verified (revision 3e8a350b3a935d68fe70bdf379692500abc9ff51), split tra
- Datasetclm-project/jev-boost-kn3-webinstruct-mmlupro-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn3-webinstruct-mmlupro: Qwen3.6-35B-A3B-FP8 state/action embeddings TIGER-Lab/WebInstruct-verified (revision 3e8a350b3a935d68fe70bdf379692500abc9ff51), split
- Datasetclm-project/jev-boost-kn3-spartqa-yesno-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn3-spartqa-yesno: Qwen3.6-35B-A3B-FP8 state/action embeddings tasksource/spartqa-yn (revision 150c819e88bba1859b8a763fd40d3a15033adc58), split train, converte
- Datasetclm-project/jev-boost-kn3-logic-entailment-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn3-logic-entailment: Qwen3.6-35B-A3B-FP8 state/action embeddings tasksource/logical-entailment (revision c2c5d47d66b01e8afa99f8da51ed0351a8c69dd2), split trai
- Datasetclm-project/jev-boost-kn3-logicnli-validity-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn3-logicnli-validity: Qwen3.6-35B-A3B-FP8 state/action embeddings tasksource/LogicNLI (revision ac9d30ac7351c5f9bb31970b7ae82182b8b4f5e8), split train, conver
- Datasetclm-project/jev-boost-kn3-strategyqa-yesno-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn3-strategyqa-yesno: Qwen3.6-35B-A3B-FP8 state/action embeddings tasksource/strategy-qa (revision f0e71799fcfdee70d1618765168904370cfec189), split train, conv
- Datasetclm-project/jev-boost-kn3-com2sense-plausible-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn3-com2sense-plausible: Qwen3.6-35B-A3B-FP8 state/action embeddings tasksource/com2sense (revision 99d1b25627a0b38202d5744c0c71baeef7b648e2), split train, con
- Datasetclm-project/jev-boost-kn3-knk-truth-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn3-knk-truth: Qwen3.6-35B-A3B-FP8 state/action embeddings K-and-K/knights-and-knaves (revision 2f68547989981b1af37cb3dde5fdefa847aa8619), split train, convert
- Tool@agentistics/mcp
Agentistics MCP server — Claude Code analytics for Claude Desktop and Claude Code
- Datasetclm-project/jev-boost-kn3-rtp-sata-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn3-rtp-sata: Qwen3.6-35B-A3B-FP8 state/action embeddings allenai/real-toxicity-prompts (revision f21629712ffd6a3d13a54fd2807ccd521c55ef74), split train, conve
- Datasetclm-project/jev-boost-kn3-multirc-sata-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn3-multirc-sata: Qwen3.6-35B-A3B-FP8 state/action embeddings aps/super_glue (revision 3de24cf8022e94f4ee4b9d55a6f539891524d646), split validation, config mult
- Datasetclm-project/jev-boost-kn3-biotech-events-sata-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn3-biotech-events-sata: Qwen3.6-35B-A3B-FP8 state/action embeddings knowledgator/events_classification_biotech (revision 96f7d85248a14441fefb613998e1484a1dbb2
- Datasetclm-project/jev-boost-kn3-eurovoc-sata-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: kn3-eurovoc-sata: Qwen3.6-35B-A3B-FP8 state/action embeddings EuropeanParliament/EuroVoc (revision ffc0177539e3f6b5edd8abb5f98fb7e8df652c6f), split train, conv
- Model
- Model
- Toolnodejs-store
Multi-backend data layer for Node.js (MongoDB, MySQL, SQLite, PostgreSQL, local disk): pure JSON schemas, GQL tree queries compiled to a single native query, GROUP BY/HAV
- Toolbunqueue
High-performance job queue for Bun and AI agents. Memory or one-file SQLite, optional PostgreSQL 15–18 multi-broker persistence, cron, retries, DLQ, and MCP.
- Datasetclm-project/jev-boost-ar5-classical-midi-score-chords-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: ar5-classical-midi-score-chords: Qwen3.6-35B-A3B-FP8 state/action embeddings drengskapur/midi-classical-music (revision 29c3079aedf539041b1cfd8ffd00d5d508681f1
- Datasetclm-project/jev-boost-ar4-polis-group-consensus-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: ar4-polis-group-consensus: Qwen3.6-35B-A3B-FP8 state/action embeddings jonaskg/open-deliberation (revision 3fe511bfd6ca9a37d822c49c1aa9e259d49c2569), split tra
- Model
- Datasetclm-project/jev-boost-ar4-hooktheory-leadsheet-chords-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: ar4-hooktheory-leadsheet-chords: Qwen3.6-35B-A3B-FP8 state/action embeddings MMR-Lab/SheetSage-A2LS-subset100 (revision 4d041baadb89e4981def2a2a07437376dd031e4
- Dataset
- Datasetclm-project/jev-boost-ar4-ugsheets-lyrics-midi-chords-qwen3.6-35b-a3b-fp8-embeddings
Jev boost: ar4-ugsheets-lyrics-midi-chords: Qwen3.6-35B-A3B-FP8 state/action embeddings jammai/chords_and_lyrics (revision 1044afa78603a0ac47c8b7f2673eb012dac50fe7), spli
Headlines and links come from the labs' own blogs, Hacker News and the GDELT news index, and from the sources our index already follows. We show a short excerpt and always link to the original. A launch reported by a news site rather than the lab is marked “Reported” until the lab itself announces it.