whatmodel.sh / decision / baselabs mammoth-1 (27B)

baselabs mammoth-1 (27B)

+ contribute

baselabs mammoth-1 (27B) is an open-weights decision model from baselabs, released 21 October 2026. Its published size is 27.36B parameters. Licence: mit.

Catalogue identity notes

Exact Decision Index tested engine mammoth-1; reported model baselabs mammoth-1 (27B). Type: full fine-tune; technique: autoregressive. Published checkpoint/base reference: 9d1962f193e48a68cf0f3598dc7fb4462fd23ad9. Comparison entry; no current run command, API price or speed inferred from the benchmark. Current repository metadata checked separately; an unpinned alias is not evidence that today’s weights reproduce this historical run.

Model information sources

Release date ↗ · Weights licence ↗ · Parameter count ↗

Publisher parameter count; includes the served backbone, not just adapter size.

74.8±1.3#2 of 355 decision modelsfrom 60 benchmark results across 4 sources

$ how it compares

The decision models ranked either side of baselabs mammoth-1 (27B), out of 357 with a score, then more from baselabs.

#	model	score	where	released
1	Jeffan-31b-1710 tattrongvu	76±2	model files
2	baselabs mammoth-1 (27B) baselabs	75±1	model files	2026-10-21
3	Sieve 31B v3 Sthānika AI	74±2	model files	2026-10-27
4	Fastino GLiDE 1.2 Fastino	73±1	model files	2026-10-22
5	KnowLine 31B Gen1 PelaAI	73±2	model files	2026-10-24
	more from baselabs
24	baselabs elephant-1 (9B) baselabs	69±1	model files	2026-10-21
190	baselabs joey-1 (2B) baselabs	52±1	api price unknown

see all decision models · find the ones that fit your machine

$ what it's good at

tools & automation▁▃▁▁▁▂▃▄▆██▆ #5 of 250 · 3.2% of its score · 6 tests

Beats Jeffan-31b-1710 here, though they rank higher overall. Trails Clef-flash, Clef-omni, Clef here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: API-Bank (accuracy) submitted	89.0% ✎	▁▁▁▁▁▁▁▂▃▂█▃ beats 229/236	easy · avg 70%	▇ 0.9%	sharp · maxed out · noisy
Decision Index 0.4.1: ToolRet (nDCG@10) submitted	70.4% ✎	▁▁▁▁▁·▁▁▁▄█▃ beats 234/246	medium · avg 60%	▇ 0.9%	blunt
Decision Index 0.4.1: When2Call (accuracy) submitted	91.7% ✎	▁▁▁▂▃▅▅▆█▇█▃ beats 236/246	easy · avg 72%	▇ 0.6%	noisy
Decision Index 0.4.1: Home appliance simulator (case exact accuracy) submitted	100.0% ✎	█▁▁▃▄▄▆▄▁▂▁▃ beats 238/243	medium · avg 37%	▇ 0.5%	sharp · noisy
Decision Index 0.4.1: BFCL (case exact accuracy) submitted	98.2% ✎	▁▂▁▁▁▂▂▄▅▆█▃ beats 242/246	very easy · avg 92%	▇ 0.4%	sharp · maxed out · noisy
Decision Index 0.4.1: RouterBench (selected quality (quality objective)) submitted	79.3% ✎	no other model yet		—	failed run: Publisher excludes this protocol because its prompt reveals GPT-4 as the best route. Evidence: https://huggingface.co/spaces/multimodalart/jev-decision-index/blob/c130faf65f82e81d3bdefb01e758c712dc5c5199/data/index.json

retrieval & classification▁▁▁·▁▁▁▁▃▃█▅ #3 of 258 · 14% of its score · 13 tests

Beats Jeffan-31b-1710 here, though they rank higher overall. Trails StartLux-Decision v1.1 27B, Sieve 31B v3 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: BRIGHT (nDCG@10) submitted	48.1% ✎	▁▁▁▁▁▁▁▂▂▄▇█ beats 198/246	medium · avg 38%	▇ 2.6%	blunt · at its level
System One Bench: MIRACL pairwise (Arabic)	90.0% ✎	▂···▁··▃▄▂█▇ beats 19/29	easy · avg 84%	▇ 2.6%	typical
Decision Index 0.4.1: Amazon ESCI (macro-F1) submitted	64.6% ✎	▁▁▁▁▁▁▂▂▃▃█▅ beats 245/246	medium · avg 47%	▇ 2.2%	at its level
System One Bench: MASSIVE intent (Arabic)	91.0% ✎	▄·▁·▁▃▃▃█▄▇▇ beats 27/29	easy · avg 76%	▇ 2.0%	typical
System One Bench: MIRACL rerank (Arabic)	89.0% ✎	▁▂·▁▁·▂·▄▄█▇ beats 24/29	easy · avg 79%	▇ 1.4%	typical
System One Bench: MASSIVE scenario (Arabic)	75.0% ✎	▃▁·▄▃▂█▃▁▁·▃ beats 23/29	medium · avg 60%	▇ 1.1%	typical
Decision Index 0.4.1: HoVer (accuracy) submitted	90.8% ✎	▆▆▇▅▄▄▃▇▄█▆▄ beats 246/246	easy · avg 74%	▇ 0.4%	blunt
Decision Index 0.4.1: PhishNChips phishing decisions (accuracy) submitted	97.1% ✎	█▅▅▄▃▂▁▁▃▁·▁ beats 241/245	easy · avg 70%	▇ 0.4%	noisy

5 more tests

test	result	vs the others in its run	difficulty	counts	why
System One Bench: AG News (English)	92.0% ✎	▁···▅▁▅▂▆█▂▅ beats 29/29	easy · avg 88%	▇ 0.3%	blunt
Decision Index 0.4.1: BANKING77 (macro-F1) submitted	93.9% ✎	▁·▁▁▁▁▁▁▂█▇█ beats 235/238	easy · avg 79%	▇ 0.2%	noisy
Decision Index 0.4.1: CLINC150+OOS (macro-F1) submitted	97.6% ✎	▁▁▁▁▁▁▁▂▃▆█▄ beats 238/238	easy · avg 83%	▇ 0.2%	sharp · maxed out · noisy
System One Bench: Banking77 (English)	94.5% ✎	█▅▆█▅▃▃▃···▂ beats 23/26	very easy · avg 87%	▇ 0.0%	blunt · noisy
Decision Index 0.4.1: SGD/SGD-X (macro-F1) submitted	42.7% ✎	no other model yet		—	failed run: Publisher reports a builder bug exposing dialogue turns after the classified turn; awaiting corrected rerun. Evidence: https://huggingface.co/spaces/multimodalart/jev-decision-index/blob/c130faf65f82e81d3bdefb01e758c712dc5c5199/data/index.json

language understanding▂▂▂▃▂▅▆▃█▆▄▃ #1 of 256 · 22% of its score · 18 tests

Beats Jeffan-31b-1710 here, though they rank higher overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: ANLI (macro-F1) submitted	82.9% ✎	▁▂▃▅▅█▃····▁ beats 245/246	medium · avg 60%	▇ 3.7%	typical
Decision Index 0.4.1: NLI4CT (macro-F1) submitted	87.9% ✎	▁▁▁▂▂▃▃▄▇▇█▄ beats 245/246	easy · avg 74%	▇ 3.4%	typical
Decision Index 0.4.1: iSarcasmEval (Sarcasm F1 · track A, English) submitted	68.1% ✎	▁·▁·▁▁▂▄▅█▇▄ beats 219/246	medium · avg 44%	▇ 2.1%	at its level
System One Bench: BoolQ (English)	92.0% ✎	▁▁▁▃·▁▃▃▇▇█▅ beats 27/29	easy · avg 86%	▇ 1.9%	typical
Decision Index 0.4.1: ACOS (per-review F1) submitted	68.0% ✎	▁··▂▃▃▅▇█▇▃▃ beats 246/246	hard · avg 22%	▇ 1.6%	at its level · noisy
System One Bench: SST-5 (English)	59.0% ✎	▃·▁▁▃·▇▄▄▄█▅ beats 21/29	medium · avg 51%	▇ 1.4%	blunt · at its level
System One Bench: OSACT4 offensive (Arabic)	91.5% ✎	▁······▁▃█▇▄ beats 20/29	easy · avg 84%	▇ 1.3%	sharp
Decision Index 0.4.1: HellaSwag (accuracy) submitted	98.5% ✎	▂▄▂▂▄▇▃▇█▅▃▃ beats 244/246	very easy · avg 85%	▇ 1.3%	sharp · maxed out

10 more tests

test	result	vs the others in its run	difficulty	counts	why
System One Bench: SST-2 (English)	96.0% ✎	▁·▃▁▂▂▃▅█·▂▁ beats 25/29	easy · avg 92%	▇ 1.0%	maxed out
Decision Index 0.4.1: VAST (macro-F1) submitted	75.4% ✎	▁▁▁▂▅▄▄▂▄▆█▄ beats 175/246	medium · avg 62%	▇ 0.8%	typical
Decision Index 0.4.1: ContractNLI (macro-F1) submitted	85.6% ✎	▁▁▁▁▁▁▁▂▂▇▇█ beats 237/246	easy · avg 71%	▇ 0.6%	noisy
System One Bench: XNLI Arabic (Arabic)	78.7% ✎	▄▁▃▁▄▇█▅▃▁·▃ beats 22/29	medium · avg 68%	▇ 0.6%	at its level
System One Bench: XNLI French (French)	86.0% ✎	▂▅·▁▁▅▅█▃▁·▃ beats 25/29	medium · avg 76%	▇ 0.5%	blunt
System One Bench: Tweet sentiment (Arabic)	64.5% ✎	▁······▁▁▄█▄ beats 10/29	medium · avg 62%	▇ 0.5%	blunt · at its level
Decision Index 0.4.1: RAGTruth (F1 on hallucinated class) submitted	85.8% ✎	▁▁▁▁▁▁▁▁▁▃▆█ beats 238/246	easy · avg 66%	▇ 0.5%	noisy
Decision Index 0.4.1: FinEntity (macro-F1) submitted	95.4% ✎	▁···▁▁▁▁▃▆█▃ beats 237/246	easy · avg 84%	▇ 0.3%	maxed out · noisy
System One Bench: XNLI English (English)	91.3% ✎	▃·▁·▃▇█▄█▃▃▁ beats 24/29	easy · avg 85%	▇ 0.2%	blunt · noisy
System One Bench: DAIR Emotion (English)	59.0% ✎	██▃·····▁··▁ beats 20/29	medium · avg 59%	▇ 0.0%	blunt · at its level

arts & human taste▁▇▄▃▅▅█▅▄▂▁▃ #3 of 247 · 7.3% of its score · 6 tests

Beats Jeffan-31b-1710 here, though they rank higher overall. Trails vera-large, Sieve 31B v3 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: New Yorker (accuracy) submitted	87.3% ✎	▁▁▂▃▃▃▇▅█▅▃▃ beats 244/246	medium · avg 61%	▇ 2.7%	typical
Decision Index 0.4.1: Humicroedit (accuracy) submitted	69.4% ✎	▁·▁▃▂▂▅█▄▃▁▁ beats 241/246	medium · avg 60%	▇ 1.9%	blunt · at its level
Decision Index 0.4.1: cfcolor (accuracy) submitted	67.7% ✎	▁········▁▇█ beats 231/246	medium · avg 60%	▇ 1.1%	blunt · at its level
Decision Index 0.4.1: POP909-CL (accuracy) submitted	77.7% ✎	▄▅▅▃▆▅▃▅▂▅▃█ beats 205/237	hard · avg 21%	▇ 0.9%	sharp · at its level · noisy
Decision Index 0.4.1: BPoMP (accuracy) submitted	99.1% ✎	▁█▄▄▅█▆▅▄▁▃▁ beats 244/246	easy · avg 83%	▇ 0.5%	sharp · maxed out · noisy
Decision Index 0.4.1: Habermas Machine (accuracy) submitted	63.4% ✎	▁▂▄█▇▅▁▁▄▄▁▁ beats 232/246	medium · avg 47%	▇ 0.3%	blunt · at its level

knowledge & reasoning▁▂▂▂▂▃▃▅▅▆█▃ #3 of 263 · 42% of its score · 16 tests

Beats Jeffan-31b-1710 here, though they rank higher overall. Trails Jev 1.13, Perplexity Decider v1.1 (27B) here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Etsabary · 1342 text decisions	84.9% ✎	▁··▂·▃▅▁▃▂▂█ beats 24/24	medium · avg 47%	▇▇▇▇ 10%	sharp
Decision Index 0.4.1: MMLU-Pro (accuracy) submitted	75.0% ✎	▄▄▃▃▂▄▅▇██▅▁ beats 239/246	medium · avg 46%	▇▇▇ 7.4%	sharp
CONFLICTS · source conflict types	65.8% ✎	▃▃··▅▃···▃▃█ beats 6/9	medium · avg 45%	▇▇ 3.8%	at its level
Decision Index 0.4.1: BBH (accuracy) submitted	96.2% ✎	▄▆▄▄█▇▇▄▃▁▁▃ beats 246/246	easy · avg 66%	▇ 3.6%	typical
Decision Index 0.4.1: GPQA Diamond (accuracy) submitted	59.2% ✎	▂▆▅▅▆█▆▃▁··▁ beats 244/246	medium · avg 39%	▇ 3.4%	blunt · at its level
Decision Index 0.4.1: MMLU (accuracy) submitted	90.0% ✎	▁▄▄▃▃▂▅▇▅█▇▃ beats 242/246	easy · avg 70%	▇ 3.3%	sharp
Decision Index 0.4.1: GSM8K (accuracy (4- and 10-choice)) submitted	88.5% ✎	▁··▄█▆▆▆▅▅▃▃ beats 242/246	medium · avg 45%	▇ 2.4%	sharp
Decision Index 0.4.1: SATA-Bench (case exact accuracy) submitted	62.2% ✎	▁▁▁▁▁▁▁▂▅█▂▃ beats 246/246	hard · avg 22%	▇ 1.9%	at its level

8 more tests

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: WinoGrande (accuracy) submitted	96.1% ✎	█▅█▅▄█▆▄▃▃▃▁ beats 242/246	easy · avg 78%	▇ 1.6%	maxed out
Decision Index 0.4.1: MuSR (accuracy) submitted	91.9% ✎	▁▁▂█▅▄▁▁▁▁▁▃ beats 245/246	medium · avg 60%	▇ 1.1%	blunt · at its level
Decision Index 0.4.1: ChessBench (accuracy) submitted	33.1% ✎	▄▇▄▄▄▆▄█▄▁▃▁ beats 243/246	hard · avg 16%	▇ 0.9%	blunt · at its level
Decision Index 0.4.1: CRUXEval (accuracy) submitted	95.4% ✎	██▄▃▃▄▁▃···▁ beats 244/246	medium · avg 58%	▇ 0.8%	noisy
Decision Index 0.4.1: ARC-Challenge (accuracy) submitted	97.6% ✎	▂▂▂▂▂▂▁▃▄▆█▆ beats 229/246	very easy · avg 90%	▇ 0.7%	sharp · maxed out
Decision Index 0.4.1: CLadder (accuracy) submitted	95.7% ✎	▁███▆▅▃▃▁▁▃▁ beats 242/246	easy · avg 71%	▇ 0.7%	typical
Decision Index 0.4.1: ARC-Easy (accuracy) submitted	99.4% ✎	▁▁▂▁▂▂▂▂▃▆█▅ beats 241/246	very easy · avg 96%	▇ 0.2%	sharp · maxed out · noisy
Decision Index 0.4.1: HLE (accuracy) submitted	19.6% ✎	▁▁▂▃▆█▇▅▅▃▁▃ beats 244/246	very hard · avg 13%	▇ 0.0%	blunt

workflows▁▁▁▂▁▂▂▆█▄▁▁ #11 of 114 * · 2.8% of its score · 1 test

Trails Microsoft-Decision-1, bosun-v3.1-0.6b (DecisionBench aaa9dd06), Mercury Decide (paid endpoint) here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
System One Bench: Typed decisions (English)	76.4% ✎	▁▁·▁▁▁▅▄▇██▁ beats 28/29	medium · avg 63%	▇ 2.8%	blunt

other tests2 not in a task · 8.9% of its score

test	result	vs the others in its run	difficulty	counts	why
Decision Index 0.4.1: private same skills (equated points)	65.9800 points ✎	▂▂▂▃▃▃▄▅▅▇█▃ beats 243/247	avg 41.206 points	▇▇ 5.8%	blunt
Decision Index 0.4.1: private new domains (equated points)	72.1400 points ✎	▃█▃▃▂▆▃▃▄▄▄▄ beats 219/247	avg 38.861 points	▇ 3.2%	blunt

$ where to run it

Download model files ↗
Published model files verified. This is the revision recorded by the benchmark. A local run command and hardware requirements have not been verified for this entry. Source · checked 2026-10-31.

Model repository and documentation: huggingface.co/baselabs ↗

$ how the score was worked out · Decision Index 0.4.1 68%, System One Bench (Ali Alhashi… 18%, Etsabary · decision benchmarks 10% and 1 more

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

source	results	share of the score	licence
Decision Index 0.4.1 · submitted; independent execution unverified	42	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 68%	CC BY 4.0
System One Bench (Ali Alhashimi)	16	▇▇▇▇▇ 18%	CC-BY-4.0
Etsabary · decision benchmarks	1	▇▇▇ 10%	MIT
Do We Need an LLM to Know Why Sources Conflict? Decision Models for Knowledge-Conflict Typing in RAG	1	▇ 4%	CC-BY-4.0
prior · pull toward the middle		▇ 1%
weighted average		= 74.8

not counted (2)

Decision Index 0.4.1	Decision Index 0.4.1: RouterBench (selected quality (quality objective)) [correct]	79.3%	failed run: Publisher excludes this protocol because its prompt reveals GPT-4 as the best route. Evidence: https://huggingface.co/spaces/multimodalart/jev-decision-index/blob/c130faf65f82e81d3bdefb01e758c712dc5c5199/data/index.json
Decision Index 0.4.1	Decision Index 0.4.1: SGD/SGD-X (macro-F1) [correct]	42.7%	failed run: Publisher reports a builder bug exposing dialogue turns after the classified turn; awaiting corrected rerun. Evidence: https://huggingface.co/spaces/multimodalart/jev-decision-index/blob/c130faf65f82e81d3bdefb01e758c712dc5c5199/data/index.json

$ improve this page

Update model information Add a place to run it Add a test result Propose a benchmark Correct a benchmark score Edit 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 2026-11-03 02:37 UTC · sources and licences · json