Fastino LabsDecision modelOpen source

GLiNER2.5-multi-Decide

PopularityRank 178 of 875

45out of 100-2 vs. last month

Summary

GLiNER2.5-multi-Decide is Fastino’s multilingual 287-million-parameter classifier in the GLiNER2.5 family. You pass labels for intent, routing, sentiment, priority, or policy at call time. One forward pass returns single-label or multi-label results and ordinal scores without generated tokens.

The fastino/GLiNER2.5-multi-Decide checkpoint is based on fastino/gliner2.5-multi-v1 with an mDeBERTa-v3-base encoder. Apache-2.0 weights run locally on CPU or GPU through gliner2.

All benchmark results

Every published result with its source, test setup and comparison models.

Show all 49 published benchmark results

Measurements without matching peer values

These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.

fastino/fast-decisions: exact-match average accuracy (Show measurement, test conditions, and source)

Source value: 56.7
Score: 56.7
Metric: exact-match average accuracy
Unit: %
Category: fastino/fast-decisions
Direction: Higher is better
Sample: 5100
Source type: vendor-reported
Evaluator: Fastino GLiNER2.5-multi-Decide model card
Status: active
Retrieved at: 2026-10-04
Methodology: Vendor-reported result. Language=English; Domains=17; Held-out examples per domain=300; Same texts and candidate labels=yes; Sample size=5100

Source Fastino GLiNER2.5-multi-Decide model card

MASSIVE 1.1 test, scenario classification, de, prompt mode, accuracy (Show measurement, test conditions, and source)

Source value: 48.5
Score: 48.5
Metric: Accuracy: share of answers equal to gold; unanswered rows count as wrong.
Unit: %
Benchmark version: BKS-Lab lab-bench route1, run 1; MASSIVE 1.1 test; written 2026-10-09
Direction: Higher is better
Sample: 2974
Source type: independent-evaluator
Evaluator: BKS-Lab lab-bench route1 score table
Status: active
Retrieved at: 2026-10-24
Methodology: Independent evaluation. Exact arm gliner2.5-multi-decide, model path fastino/GLiNER2.5-multi-Decide (multilingual Decide); gliner2 2.0.0 and PyTorch 2.11.0+cu128 on the local RTX 4090 workstation. One single-label task per question; the state is the named utterance section; label and description are the readable option text and the predicted label maps back to its option key. prompt mode: instruction is in the task prompt.
Context: One full run only; the report publishes no confidence interval or repeated-run uncertainty. The method says each GLiNER row stores the Hugging Face snapshot commit SHA, but the reviewed aggregate score page does not expose its value. The public test data were not checked for training exposure; Fastino's own Decide training suite may include MASSIVE. Fast-Decisions is Fastino-built and described as the classification suite behind GLiNER2.5-Decide, so it is a familiar/possibly training-adjacent cohort, not a neutral generalization test.

Source BKS-Lab lab-bench route1 score table

MASSIVE 1.1 test, scenario classification, de, prompt mode, macro-F1 (Show measurement, test conditions, and source)

Source value: 46.4
Score: 46.4
Metric: Macro-F1: mean F1 over labels seen in gold or prediction.
Unit: %
Benchmark version: BKS-Lab lab-bench route1, run 1; MASSIVE 1.1 test; written 2026-10-09
Direction: Higher is better
Sample: 2974
Source type: independent-evaluator
Evaluator: BKS-Lab lab-bench route1 score table
Status: active
Retrieved at: 2026-10-24
Methodology: Independent evaluation. Exact arm gliner2.5-multi-decide, model path fastino/GLiNER2.5-multi-Decide (multilingual Decide); gliner2 2.0.0 and PyTorch 2.11.0+cu128 on the local RTX 4090 workstation. One single-label task per question; the state is the named utterance section; label and description are the readable option text and the predicted label maps back to its option key. prompt mode: instruction is in the task prompt.
Context: One full run only; the report publishes no confidence interval or repeated-run uncertainty. The method says each GLiNER row stores the Hugging Face snapshot commit SHA, but the reviewed aggregate score page does not expose its value. The public test data were not checked for training exposure; Fastino's own Decide training suite may include MASSIVE. Fast-Decisions is Fastino-built and described as the classification suite behind GLiNER2.5-Decide, so it is a familiar/possibly training-adjacent cohort, not a neutral generalization test.

Source BKS-Lab lab-bench route1 score table

MASSIVE 1.1 test, scenario classification, de, intext mode, accuracy (Show measurement, test conditions, and source)

Source value: 46.7
Score: 46.7
Metric: Accuracy: share of answers equal to gold; unanswered rows count as wrong.
Unit: %
Benchmark version: BKS-Lab lab-bench route1, run 1; MASSIVE 1.1 test; written 2026-10-09
Direction: Higher is better
Sample: 2974
Source type: independent-evaluator
Evaluator: BKS-Lab lab-bench route1 score table
Status: active
Retrieved at: 2026-10-24
Methodology: Independent evaluation. Exact arm gliner2.5-multi-decide, model path fastino/GLiNER2.5-multi-Decide (multilingual Decide); gliner2 2.0.0 and PyTorch 2.11.0+cu128 on the local RTX 4090 workstation. One single-label task per question; the state is the named utterance section; label and description are the readable option text and the predicted label maps back to its option key. intext mode: instruction is appended to the passage as Question: ....
Context: One full run only; the report publishes no confidence interval or repeated-run uncertainty. The method says each GLiNER row stores the Hugging Face snapshot commit SHA, but the reviewed aggregate score page does not expose its value. The public test data were not checked for training exposure; Fastino's own Decide training suite may include MASSIVE. Fast-Decisions is Fastino-built and described as the classification suite behind GLiNER2.5-Decide, so it is a familiar/possibly training-adjacent cohort, not a neutral generalization test.

Source BKS-Lab lab-bench route1 score table

MASSIVE 1.1 test, scenario classification, de, intext mode, macro-F1 (Show measurement, test conditions, and source)

Source value: 46.9
Score: 46.9
Metric: Macro-F1: mean F1 over labels seen in gold or prediction.
Unit: %
Benchmark version: BKS-Lab lab-bench route1, run 1; MASSIVE 1.1 test; written 2026-10-09
Direction: Higher is better
Sample: 2974
Source type: independent-evaluator
Evaluator: BKS-Lab lab-bench route1 score table
Status: active
Retrieved at: 2026-10-24
Methodology: Independent evaluation. Exact arm gliner2.5-multi-decide, model path fastino/GLiNER2.5-multi-Decide (multilingual Decide); gliner2 2.0.0 and PyTorch 2.11.0+cu128 on the local RTX 4090 workstation. One single-label task per question; the state is the named utterance section; label and description are the readable option text and the predicted label maps back to its option key. intext mode: instruction is appended to the passage as Question: ....
Context: One full run only; the report publishes no confidence interval or repeated-run uncertainty. The method says each GLiNER row stores the Hugging Face snapshot commit SHA, but the reviewed aggregate score page does not expose its value. The public test data were not checked for training exposure; Fastino's own Decide training suite may include MASSIVE. Fast-Decisions is Fastino-built and described as the classification suite behind GLiNER2.5-Decide, so it is a familiar/possibly training-adjacent cohort, not a neutral generalization test.

Source BKS-Lab lab-bench route1 score table

MASSIVE 1.1 test, intent classification (gold-scenario oracle), de, prompt mode, accuracy (Show measurement, test conditions, and source)

Source value: 63.8
Score: 63.8
Metric: Accuracy: share of answers equal to gold; unanswered rows count as wrong.
Unit: %
Benchmark version: BKS-Lab lab-bench route1, run 1; MASSIVE 1.1 test; written 2026-10-09
Direction: Higher is better
Sample: 2974
Source type: independent-evaluator
Evaluator: BKS-Lab lab-bench route1 score table
Status: active
Retrieved at: 2026-10-24
Methodology: Independent evaluation. Exact arm gliner2.5-multi-decide, model path fastino/GLiNER2.5-multi-Decide (multilingual Decide); gliner2 2.0.0 and PyTorch 2.11.0+cu128 on the local RTX 4090 workstation. One single-label task per question; the state is the named utterance section; label and description are the readable option text and the predicted label maps back to its option key. prompt mode: instruction is in the task prompt.
Context: One full run only; the report publishes no confidence interval or repeated-run uncertainty. The method says each GLiNER row stores the Hugging Face snapshot commit SHA, but the reviewed aggregate score page does not expose its value. The public test data were not checked for training exposure; Fastino's own Decide training suite may include MASSIVE. Fast-Decisions is Fastino-built and described as the classification suite behind GLiNER2.5-Decide, so it is a familiar/possibly training-adjacent cohort, not a neutral generalization test.

Source BKS-Lab lab-bench route1 score table

MASSIVE 1.1 test, intent classification (gold-scenario oracle), de, prompt mode, macro-F1 (Show measurement, test conditions, and source)

Source value: 57.2
Score: 57.2
Metric: Macro-F1: mean F1 over labels seen in gold or prediction.
Unit: %
Benchmark version: BKS-Lab lab-bench route1, run 1; MASSIVE 1.1 test; written 2026-10-09
Direction: Higher is better
Sample: 2974
Source type: independent-evaluator
Evaluator: BKS-Lab lab-bench route1 score table
Status: active
Retrieved at: 2026-10-24
Methodology: Independent evaluation. Exact arm gliner2.5-multi-decide, model path fastino/GLiNER2.5-multi-Decide (multilingual Decide); gliner2 2.0.0 and PyTorch 2.11.0+cu128 on the local RTX 4090 workstation. One single-label task per question; the state is the named utterance section; label and description are the readable option text and the predicted label maps back to its option key. prompt mode: instruction is in the task prompt.
Context: One full run only; the report publishes no confidence interval or repeated-run uncertainty. The method says each GLiNER row stores the Hugging Face snapshot commit SHA, but the reviewed aggregate score page does not expose its value. The public test data were not checked for training exposure; Fastino's own Decide training suite may include MASSIVE. Fast-Decisions is Fastino-built and described as the classification suite behind GLiNER2.5-Decide, so it is a familiar/possibly training-adjacent cohort, not a neutral generalization test.

Source BKS-Lab lab-bench route1 score table

MASSIVE 1.1 test, intent classification (gold-scenario oracle), de, intext mode, accuracy (Show measurement, test conditions, and source)

Source value: 63.3
Score: 63.3
Metric: Accuracy: share of answers equal to gold; unanswered rows count as wrong.
Unit: %
Benchmark version: BKS-Lab lab-bench route1, run 1; MASSIVE 1.1 test; written 2026-10-09
Direction: Higher is better
Sample: 2974
Source type: independent-evaluator
Evaluator: BKS-Lab lab-bench route1 score table
Status: active
Retrieved at: 2026-10-24
Methodology: Independent evaluation. Exact arm gliner2.5-multi-decide, model path fastino/GLiNER2.5-multi-Decide (multilingual Decide); gliner2 2.0.0 and PyTorch 2.11.0+cu128 on the local RTX 4090 workstation. One single-label task per question; the state is the named utterance section; label and description are the readable option text and the predicted label maps back to its option key. intext mode: instruction is appended to the passage as Question: ....
Context: One full run only; the report publishes no confidence interval or repeated-run uncertainty. The method says each GLiNER row stores the Hugging Face snapshot commit SHA, but the reviewed aggregate score page does not expose its value. The public test data were not checked for training exposure; Fastino's own Decide training suite may include MASSIVE. Fast-Decisions is Fastino-built and described as the classification suite behind GLiNER2.5-Decide, so it is a familiar/possibly training-adjacent cohort, not a neutral generalization test.

Source BKS-Lab lab-bench route1 score table

MASSIVE 1.1 test, intent classification (gold-scenario oracle), de, intext mode, macro-F1 (Show measurement, test conditions, and source)

Source value: 55.3
Score: 55.3
Metric: Macro-F1: mean F1 over labels seen in gold or prediction.
Unit: %
Benchmark version: BKS-Lab lab-bench route1, run 1; MASSIVE 1.1 test; written 2026-10-09
Direction: Higher is better
Sample: 2974
Source type: independent-evaluator
Evaluator: BKS-Lab lab-bench route1 score table
Status: active
Retrieved at: 2026-10-24
Methodology: Independent evaluation. Exact arm gliner2.5-multi-decide, model path fastino/GLiNER2.5-multi-Decide (multilingual Decide); gliner2 2.0.0 and PyTorch 2.11.0+cu128 on the local RTX 4090 workstation. One single-label task per question; the state is the named utterance section; label and description are the readable option text and the predicted label maps back to its option key. intext mode: instruction is appended to the passage as Question: ....
Context: One full run only; the report publishes no confidence interval or repeated-run uncertainty. The method says each GLiNER row stores the Hugging Face snapshot commit SHA, but the reviewed aggregate score page does not expose its value. The public test data were not checked for training exposure; Fastino's own Decide training suite may include MASSIVE. Fast-Decisions is Fastino-built and described as the classification suite behind GLiNER2.5-Decide, so it is a familiar/possibly training-adjacent cohort, not a neutral generalization test.

Source BKS-Lab lab-bench route1 score table

MASSIVE 1.1 test, intent classification, 2+ options, de, prompt mode, accuracy (Show measurement, test conditions, and source)

Source value: 58.9
Score: 58.9
Metric: Accuracy: share of answers equal to gold; unanswered rows count as wrong.
Unit: %
Benchmark version: BKS-Lab lab-bench route1, run 1; MASSIVE 1.1 test, intents restricted to 2+ choices; written 2026-10-09
Direction: Higher is better
Sample: 2622
Source type: independent-evaluator
Evaluator: BKS-Lab lab-bench route1 score table
Status: active
Retrieved at: 2026-10-24
Methodology: Independent evaluation. Exact arm gliner2.5-multi-decide, model path fastino/GLiNER2.5-multi-Decide (multilingual Decide); gliner2 2.0.0 and PyTorch 2.11.0+cu128 on the local RTX 4090 workstation. One single-label task per question; the state is the named utterance section; label and description are the readable option text and the predicted label maps back to its option key. prompt mode: instruction is in the task prompt.
Context: One full run only; the report publishes no confidence interval or repeated-run uncertainty. The method says each GLiNER row stores the Hugging Face snapshot commit SHA, but the reviewed aggregate score page does not expose its value. The public test data were not checked for training exposure; Fastino's own Decide training suite may include MASSIVE. Fast-Decisions is Fastino-built and described as the classification suite behind GLiNER2.5-Decide, so it is a familiar/possibly training-adjacent cohort, not a neutral generalization test.

Source BKS-Lab lab-bench route1 score table

MASSIVE 1.1 test, intent classification, 2+ options, de, prompt mode, macro-F1 (Show measurement, test conditions, and source)

Source value: 54.9
Score: 54.9
Metric: Macro-F1: mean F1 over labels seen in gold or prediction.
Unit: %
Benchmark version: BKS-Lab lab-bench route1, run 1; MASSIVE 1.1 test, intents restricted to 2+ choices; written 2026-10-09
Direction: Higher is better
Sample: 2622
Source type: independent-evaluator
Evaluator: BKS-Lab lab-bench route1 score table
Status: active
Retrieved at: 2026-10-24
Methodology: Independent evaluation. Exact arm gliner2.5-multi-decide, model path fastino/GLiNER2.5-multi-Decide (multilingual Decide); gliner2 2.0.0 and PyTorch 2.11.0+cu128 on the local RTX 4090 workstation. One single-label task per question; the state is the named utterance section; label and description are the readable option text and the predicted label maps back to its option key. prompt mode: instruction is in the task prompt.
Context: One full run only; the report publishes no confidence interval or repeated-run uncertainty. The method says each GLiNER row stores the Hugging Face snapshot commit SHA, but the reviewed aggregate score page does not expose its value. The public test data were not checked for training exposure; Fastino's own Decide training suite may include MASSIVE. Fast-Decisions is Fastino-built and described as the classification suite behind GLiNER2.5-Decide, so it is a familiar/possibly training-adjacent cohort, not a neutral generalization test.

Source BKS-Lab lab-bench route1 score table

MASSIVE 1.1 test, intent classification, 2+ options, de, intext mode, accuracy (Show measurement, test conditions, and source)

Source value: 58.4
Score: 58.4
Metric: Accuracy: share of answers equal to gold; unanswered rows count as wrong.
Unit: %
Benchmark version: BKS-Lab lab-bench route1, run 1; MASSIVE 1.1 test, intents restricted to 2+ choices; written 2026-10-09
Direction: Higher is better
Sample: 2622
Source type: independent-evaluator
Evaluator: BKS-Lab lab-bench route1 score table
Status: active
Retrieved at: 2026-10-24
Methodology: Independent evaluation. Exact arm gliner2.5-multi-decide, model path fastino/GLiNER2.5-multi-Decide (multilingual Decide); gliner2 2.0.0 and PyTorch 2.11.0+cu128 on the local RTX 4090 workstation. One single-label task per question; the state is the named utterance section; label and description are the readable option text and the predicted label maps back to its option key. intext mode: instruction is appended to the passage as Question: ....
Context: One full run only; the report publishes no confidence interval or repeated-run uncertainty. The method says each GLiNER row stores the Hugging Face snapshot commit SHA, but the reviewed aggregate score page does not expose its value. The public test data were not checked for training exposure; Fastino's own Decide training suite may include MASSIVE. Fast-Decisions is Fastino-built and described as the classification suite behind GLiNER2.5-Decide, so it is a familiar/possibly training-adjacent cohort, not a neutral generalization test.

Source BKS-Lab lab-bench route1 score table

Page 1 of 5

Popularity compared

How often the model is searched for, used and mentioned, compared with every model of its category.

Popularity index

0 to 100, higher is more popular

20 of 37 models

AutoTrust JEV-27B-VL70

Clef51

GLiNER2.5-multi-Decide45

Decider 4B43

Jev43

pplx-decider-v1.1-27b42

Intern-Decision-4B40

Laya Typed Decisions40

Kev 9B40

pplx-decider-v1-27b39

Decider 0.8B39

Julia 139

Intern-Decision-0.8B38

GLiNER2.5-multi-Decide

Decider 4B

Jev

pplx-decider-v1.1-27b

As of November 2026 · Source: gradually.ai popularity index

Over time

Weekly values from October 18, 2026 to November 5, 2026

GLiNER2.5-multi-DecideKev 0.8BLaya MultilingualDecider 4BJev

All specifications

Architecture

Model class: Multilingual specialist classifier for operational decisions, not a general-purpose model; Source: Model class
Checkpoint: fastino/GLiNER2.5-multi-Decide; Source: Checkpoint
Base model: fastino/gliner2.5-multi-v1, mDeBERTa-v3-base encoder; Source: Base model

Input and output

Input: Multilingual text with caller-supplied labels and optional descriptions; Source: Input

Watermarking and provenance

AI watermarking refers to invisible signals embedded in the output. Visible marks and provenance metadata are listed separately. Unknown means there is insufficient evidence for that particular field.

AI watermarking: Unknown; no source
Watermarking technique: Not known; no source

Licence and access

Access: Local CPU or GPU execution through gliner2; Source: Access
License: Apache-2.0; Source: License

Further details

Output schema: Single-label or multi-label classification and ordinal scores without generated tokens; Source: Output schema

Sources and data date

Show all 15 sources

Benchmark sources at a glance

Which independent evaluators tested this model. This is coverage, not a quality ranking.

No active result from the reviewed external evaluators can currently be attributed to this model. Other source evidence remains available below.

Without vendor claims and historical or disputed results; those stay labelled in the single values.

Type	Evidence and data date
Research status	Research date October 4, 2026. 10 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication.
Not published	non-english-heldout-results Note The public multilingual card and linked dataset expose an aggregate benchmark result but no non-English held-out score breakdown.
Not found in the inspected sources	hanno-decisionbench-exact-checkpoint-result Note The complete results tree at pinned commit a512f9ec09abac2c4ebd2c2208c92ae339a3a056 contains no exact DecisionBench run for this checkpoint. Results for GLiNER2.5-Decide, GLiNER2.5-base, GLiNER2.5-multi-base, or another sibling are not transferred.
Not published	evaluator-loaded-weight-digest Note The report does not disclose loaded checkpoint hash or revision. The matching public Hugging Face revision by lastModified date is only a chronology candidate.
Not published	test-dataset-revision Note Fixed samples and sample counts are reported, but dataset revision hashes are not pinned in the captured protocol.
Not published	unreported-confidence-intervals Note No CI is published for ECE or MASSIVE per-language accuracy; none was calculated or inferred.
Unresolved	Per-run immutable GLiNER snapshot revision Note The route1 method says each GLiNER raw row stores the Hugging Face snapshot commit SHA, but the reviewed aggregate score table does not expose the SHA value. The exact run-specific weight revision therefore remains unknown.
Not published	Confidence intervals and repeated-run uncertainty Note The route1 score report is run 1 of each arm and does not publish confidence intervals or a repeated-run uncertainty estimate. None was calculated or inferred.
Not found in the inspected sources	Exact model result in the public Decision Index 0.4.1 panel Note This index edition has no row that can be mapped to the profile repository/checkpoint exactly. This does not imply the model lacks results in other benchmarks.
Additional source	Fastino model index (retrieved October 3, 2026)
Additional source	GLiNER2.5-multi-Decide model card (retrieved October 3, 2026; October 4, 2026) · GLiNER2.5-multi-Decide model card · Editorial description reviewed October 3, 2026
Additional source	System One report, Laya AI-hosted result summary for gliner2-5-multi-decide, pinned source commit ed4544869a9480aa98db6572a21733803eadcbb8 (retrieved October 17, 2026)
Additional source	BKS-Lab lab-bench route1 score table (retrieved October 24, 2026)
Additional source	Model metadata (retrieved October 16, 2026)1 2
Additional source	Model metadata (retrieved October 17, 2026)1 2
Additional source	Model metadata (retrieved October 24, 2026)
Additional source	OpenRouter: Popularity index (derived), CC BY 4.0
Additional source	Vercel AI Gateway: Popularity index (derived), CC BY 4.0
Additional source	Ollama: cumulative downloads for the popularity index
Additional source	Replicate: cumulative runs for the popularity index
Additional source	Google Trending Now: limited feed selection for Germany and the US
Additional source	Creative Commons Attribution 4.0