Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
external benchmark
Decision Index 0.4 Tools. Every model the catalog carries a published Decision Index Tools value for, ranked by that value.
Decision Index 0.4 publishes chance-corrected tools scores for 209 evaluated systems.
120 models with a published Decision Index Tools value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.
| Rank | Model | Provider | Chance-corrected area score |
|---|---|---|---|
| 1 | BAmammoth-1 | Baseten | 83.9 |
| 2 | Perplexity | 82.4 | |
| 3 | NLFastino GLiDE no-thinking (28B) | Not listed | 82.3 |
What the benchmark measures and how its value is published, from the source’s own record.
Decision Index Tools (Decision Index 0.4 Tools) is an external benchmark published by Decision Index leaderboard. Decision Index 0.4 publishes chance-corrected tools scores for 209 evaluated systems. Its evaluation set is described by the source as tools area of the Decision Index evaluation. The source rates its difficulty as tool-use and automation decision tasks.
Each model's result is published by Decision Index leaderboard as chance-corrected area score. ModelScale prints that value to 1 decimal place, exactly as published, and never converts or rescales it. This page ranks the 120 models that carry a published value, highest value first; a model the source never scored is left out rather than ranked at zero. The source does not state whether a higher value is the better result, so neither does this page.
| 4 | BAelephant-1 | Baseten | 82.1 |
| 5 | Perplexity | 80.6 |
| 6 | NLTorchcast Decision 27B | Not listed | 78.9 |
| 7 | CAEikos-27B | caiovicentino1 | 78.5 |
| 8 | JAKev 27B | jaredpalmer | 77.5 |
| 8 | NLvLLM-SR Decision 2.0 Vega 27B | Not listed | 77.5 |
| 10 | FIJebadiah 27B | frontier-infra | 77.0 |
| 11 | FAsimple-jev · Qwen3.8-27B | Featherless AI | 75.8 |
| 12 | Perplexity | 75.7 |
| 12 | KSreflex-27b | kshetrajna12 | 75.7 |
| 14 | NLdeck31b (Gemma 4 31B) | Not listed | 75.6 |
| 15 | CLClef | Cloudflare | 75.5 |
| 16 | TAJev 1.14 | TypeSafe AI | 74.4 |
| 16 | CHQuyet-1.0-Large | chinhnc | 74.4 |
| 18 | AUJEV-27B | autotrust | 73.9 |
| 19 | BEBespoke Nimble 9B v3 | bespokelabs | 73.8 |
| 20 | NLJev | Not listed | 73.7 |
| 21 | NLJADE | Not listed | 73.6 |
| 22 | NADrex v2.0 | Nace.AI | 72.5 |
| 23 | NLJevfire | Not listed | 72.2 |
| 24 | NLXor 1.2 | Not listed | 71.6 |
| 25 | NADrex v1.5 | Nace.AI | 71.2 |
| 26 | JUXor 26B-A4B | juspay | 70.8 |
| 27 | NLdecisio 0.4.0 (Gemma 4 12B) | Not listed | 70.7 |
| 28 | MADecider chat · Gemma-4-31B | Mapika | 70.5 |
| 29 | NLDecider chat · Qwen3.6-27B | Not listed | 69.8 |
| 30 | PIBlink v0.3 26B-A4B (NVFP4) | PixilabAI | 69.4 |
| 31 | SUSurogate Rune 26B-A4B v3 | Surogate | 69.0 |
| 32 | NLdecisio 0.4.0 (Qwen3.6-35B-A3B) | Not listed | 66.9 |
| 32 | NLJoshuaSP diffusiongemma (open-jev) | Not listed | 66.9 |
| 34 | NLHopper 12B (trained) | Not listed | 66.5 |
| 35 | KIJPT-35B-A3B | kirp | 66.4 |
| 36 | AUGEV-26B-Decide | autotrust | 66.3 |
| 37 | ANAJev lora5 (Gemma 4 12B) | andyzhang232 | 66.1 |
| 38 | CLClef-Flash | Cloudflare | 66.0 |
| 39 | NLvLLM-SR Decision 2.0 Lux 9B | Not listed | 65.0 |
| 40 | ELjiwo 4B | eljiwo | 63.9 |
| 41 | NLLev1 3.7B | Not listed | 63.7 |
| 42 | AKJev-Omni | akhilaaa3 | 63.4 |
| 43 | ERWinnow-12B Q8 | Eldan Ring | 62.9 |
| 44 | EVezjev 4B s2 | everettjf | 62.6 |
| 45 | HOHopper (G) 1.2 | HopitAI | 62.2 |
| 46 | SASieve 9B | sthanika-ai | 62.1 |
| 47 | KIJPT-9B | kirp | 61.0 |
| 48 | ALJevK5 9B v0.3.3 | alibiserikbay | 60.8 |
| 49 | CLClef-Omni | Cloudflare | 60.0 |
| 50 | NLKev 9B v2 | Not listed | 58.7 |
| 51 | NLvLLM-SR Decision 2.0 Nox 4B | Not listed | 58.4 |
| 52 | IAlev | interfaze-ai | 58.0 |
| 53 | AMJevstral 8B | AmirBraham | 57.6 |
| 54 | MIJet v6.2 | michaljach | 57.0 |
| 55 | AUJEV-9B | autotrust | 56.8 |
| 56 | KIJPT-4B | kirp | 56.1 |
| 57 | internlm | 55.8 |
| 58 | TONeoHorse-Jev-4B | TokenRhythm | 55.7 |
| 59 | NLIntelif Qwen3-4B | Not listed | 54.9 |
| 60 | NLrazorback16 openjev diffusiongemma (NVFP4, vLLM) | Not listed | 53.9 |
| 61 | KPKas 4B | kpiya | 53.8 |
| 62 | ELWinnow-E4B | EldanRing | 53.5 |
| 63 | BEBespoke Nimble 9B v2 | bespokelabs | 53.0 |
| 63 | NLJevK5 v0.3 | Not listed | 53.0 |
| 65 | MAdjev | Maisa | 52.9 |
| 66 | NLRyotide Qwen v0.3.1 | Not listed | 51.7 |
| 67 | JAKev 4B r10 | jaredpalmer | 50.1 |
| 68 | SHRSI-Jev v5.0 VL 3B | shgao | 49.5 |
| 69 | DAGevva E4B | davidburhans | 48.4 |
| 70 | MAdecider-35b-a3b | Mapika | 48.3 |
| 71 | NLJobe | Not listed | 48.2 |
| 72 | MADecider 4B | Mapika | 47.7 |
| 73 | NLJev-Style 2B v3 | Not listed | 47.2 |
| 74 | NLopenvons | Not listed | 46.9 |
| 75 | NLvLLM-SR Decision 2.0 Sol 2B | Not listed | 46.6 |
| 76 | NLSolomon v1.1 | Not listed | 45.6 |
| 77 | DAGevva E2B | davidburhans | 45.0 |
| 78 | NLEXAONE-4.0-1.2B-JEV v0.3 | Not listed | 44.3 |
| 79 | TOTev1-4B-experimental | togethercomputer | 44.2 |
| 80 | OMOneJev 0.8B | OmniJev | 40.2 |
| 81 | NLSemIf | Not listed | 39.6 |
| 82 | W(metask-jev-4b | Wayfind (metask-ai) | 38.4 |
| 82 | NLthis-that 1.2 | Not listed | 38.4 |
| 84 | internlm | 36.9 |
| 85 | NLopen-jev (pngwn) | Not listed | 36.7 |
| 86 | NLmmastrac diffusiongemma (vLLM PR 57250) | Not listed | 36.6 |
| 87 | MAdecider-2b | Mapika | 35.7 |
| 88 | KIJPT-0.8B | kirp | 34.0 |
| 89 | ELjiwo 0.8B | eljiwo | 32.7 |
| 90 | LUDinah-0 | Lukitaduarte | 32.2 |
| 91 | HLBosun v3.1 1.7B | Hanno-Labs | 31.9 |
| 92 | SASieve 2B | sthanika-ai | 31.7 |
| 93 | NLSifr 0.8B v3.1 | Not listed | 31.3 |
| 94 | NLKev 0.8B r15 | Not listed | 30.6 |
| 94 | TOTev1-0.8B-experimental | togethercomputer | 30.6 |
| 96 | NLmini-jev | Not listed | 29.6 |
| 97 | internlm | 29.2 |
| 97 | NLvLLM-SR Decision 2.0 Eos 0.8B | Not listed | 29.2 |
| 99 | CLBosun v3.1 0.6B | Clause Logic | 26.5 |
| 100 | NLvLLM-SR Decision 2.0 Kai 0.6B | Not listed | 24.4 |
| 101 | NLJev-Style 0.8B v3 | Not listed | 23.8 |
| 102 | HOBekko System One v0 400M | hotchpotch | 18.2 |
| 103 | CAKodiak v0.2 1B (accuracy) | cortex-agent-llc | 12.8 |
| 104 | CLCLM-v0.1-8B | Contrastive-LM | 12.3 |
| 104 | FAGLiNER2.5-Decide | fastino | 12.3 |
| 106 | HOBekko System One v0 68M | hotchpotch | 11.3 |
| 107 | MOLavoir | moganai | 9.8 |
| 108 | MOLFM2.5-2.6B-RLCD | monotykamary | 8.1 |
| 109 | CILaya | Convai Innovations | 7.9 |
| 110 | HAQwen-2.5-1B-RLCD | harshatheg | 7.7 |
| 111 | NLVon 1.3 | Not listed | 7.6 |
| 112 | HOBekko System One v0 17M | hotchpotch | 6.8 |
| 113 | SUJulia 1 | SupersonicLabs | 6.1 |
| 114 | NLDM-JEPA 1.1 | Not listed | 5.9 |
| 115 | NLVerdict | Not listed | 5.6 |
| 116 | NOLFM2.5-350M-RLCD | notnotsamuel | 3.5 |
| 117 | NLsystem-one-gemma | Not listed | 2.5 |
| 117 | NLZev | Not listed | 2.5 |
| 119 | LMjeff | Logan Markewich | 0.0 |
| 119 | ALJevK5 Lite | alibiserikbay | 0.0 |
Evidence key: ObservedLast good
Rows are ordered by the value Decision Index leaderboard published, highest first. The source does not state whether a higher value is the better result, so this page does not either: for a benchmark that measures a rate of failure, read the table from the bottom.