Model cardKnowledgator
GLiClass large v3.0 model card
Measured on: 14-dataset zero-shot classification average (F1)
- Zero-shot F1 average
- 0.620
Author-reported; standard text-classification sets, not decisions.
GLiClass modern large v3.0
KnowledgatorOpen weights
GLiClass on a ModernBERT backbone with an 8k context window.
- typed-decisions
- —
- Metatext suite
- 0.602
- ECE untuned
- 0.234
- p50 per case
- 572 ms
not run
4/7 over baseline
four CPU cores
Specifications
- Checkpoint
- knowledgator/gliclass-modern-large-v3.0
- Publisher
- Knowledgator
- Mechanism
- Label-relevance classifier
- Backbone
- ModernBERT-large (8k context)
- Parameters
- 399M
- License
- Apache 2.0
- Runs as
- open weights, self-hosted
- Library
- gliclass
- Primitives
- Choice, Noul
- Pricing
- none — your compute
Metatext decisionsrank 8 of 21
Full leaderboard →| Configuration | Mode | Decisions | Beats baseline | ECE |
|---|---|---|---|---|
| criteria + single-hypothesis Noul (correct payload)headline | serial | 0.602 | 4/7 | 0.234 |
| criteria folded into labels | serial | 0.385 | 2/7 | 0.476 |
| default payload | serial | 0.351 | 2/7 | 0.585 |
Per-question accuracy — criteria + single-hypothesis Noul (correct payload), serial mode, 120 judgments
| Task | Question | Type | Accuracy | Baseline | Beats? |
|---|---|---|---|---|---|
| agent_traces | outcome | choice | 0.214 | 0.357 | ❌ |
| agent_traces | needs_human | noul | 0.714 | 0.714 | ❌ |
| customer_service | department | choice | 0.750 | 0.400 | ✅ |
| customer_service | refund_requested | noul | 0.950 | 0.700 | ✅ |
| customer_service | frustration | score | 0.450 | 0.500 | ❌ |
| security_incidents | severity | choice | 0.500 | 0.312 | ✅ |
| security_incidents | actively_exploited | noul | 0.750 | 0.688 | ✅ |
Published numbers
Reported by the authors or by third parties, on their own suites. Not comparable to the leaderboard above; attributed so you can check them.
Frequently asked questions
How does GLiClass modern large v3.0 compare to Jev?
On the Metatext suite GLiClass modern large v3.0 scores 0.602 against Jev's 0.863.
Is GLiClass modern large v3.0 open source?
Yes. knowledgator/gliclass-modern-large-v3.0 is released under Apache 2.0 and runs on your own hardware with the gliclass library.
Can GLiClass modern large v3.0 extract entities?
No. GLiClass modern large v3.0 answers questions about text and never returns spans; use a GLiNER-family model for extraction.