The story so far
Weights, every dated article, and repo commits on one rail.
20 repo commits since the weights landed
Teal is the weights (2026-08-18), grey ticks are repo commits.
Models › gliner2.5-small-v1 family
gliner2.5-small-v1
fastino Apache 2.0 out 2 months ago
73.9M parameters.
Verified 2026-10-25 Rank #0 in Decision models · score 0 method v1 Sources 3 Downloads 30d 24.1k · #442 of 619 tracked
Changed this week: 1 new quantized build on the Hub (4 in all).2026-10-24
Do people use it?24.1kdownloads, 30 days#442 of 619 we track · below the median+1 likes, downloads −27% this week44 likesHugging Face2026-10-24
Where it's used
Hugging Face does not say who downloads a model or with what. These are the proxies: API providers listing it, quantized builds, fine-tunes/adapters/merges, and Spaces that load it.
No API provider serves it, so its downloads are people running it themselves: 4 quantized builds and 6 Spaces.
- As a service0no API provider listed
- On own hardware4quantized builds
- Built on63 fine-tunes · 3 adapters · 0 merges
- In apps6Hugging Face Spaces
Top Spaces running it: fastino/gliner25-constrained-classification ♥ 3
Will it fit?
Weights only, an estimate: leave room for the KV cache. Method
Fits a 16 GB card at FP8
About 0.1 GB at FP8, 0.2 GB at BF16: weights only, so leave room for the KV cache.
Estimate params × bytes per param × 1.2 · how we estimate
What people built on it
10 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.
10 derivatives on the Hub, led by 4 quantized
From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.
Reference
Everything else, folded: the full tables, every source.
What the card says
1 cited sentence
Use-case evidence
2 use cases
What it is good at 2 use cases
Labels are evidence, not verdicts: measured a third-party leaderboard row for this model; author-reported a number the model author published in the card or eval files; structural a deterministic signal from the name, config or chat template, or a decision-model estimate. Each chip links to its source. All use cases.
Facts
16 fields
Listed in Decision models
Facts
- gliner2.5-small-v1 has 73.9M parameters. (safetensors metadata, observed 2026-10-24)
- License: Apache 2.0 (friction: none). Permissive; patent grant. (license file, observed 2026-10-24)
- Created 2026-08-18, last modified 2026-10-08. (HF API, observed 2026-10-24)
| Card pipeline tag | schema-extraction |
|---|---|
| Library | transformers |
| Architectures | BoundaryExtractor |
| Model type | extractor |
| Params by dtype | F32: 73,881,879 |
| Size band | S |
| MoE | no |
| Languages | en |
| License | Apache 2.0 (apache-2.0) |
| Hub snapshot | 2026-10-25 |
Servability
vLLM · SGLang
Servability
vLLM: not in registry · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: not in registry · 773cb938fe9484a8f7e7e46fb1b1fe96412c59ed
vLLM registry (7050ef881ff0835f6890a47db1859a24c65481d6) · SGLang registry (773cb938fe9484a8f7e7e46fb1b1fe96412c59ed) · Matched by architecture class name against each project's registry on main; 'supported' means the architecture is registered, not that this checkpoint was tested.
Hardware fit
3 estimates
Hardware fit ESTIMATE
Weights-only memory for gliner2.5-small-v1: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 0.2 GB · Edge (<4 GB) |
|---|---|
| FP8 (1 byte/param) | 0.1 GB · Edge (<4 GB) |
| INT4 (0.5 bytes/param) | 0.0 GB · Edge (<4 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
Decision facts
catalog
Decision-model catalog curated
| How it was made | own architecture (pre-existing encoder classifier) |
|---|---|
| Lineage bucket | O - original architecture or own base |
| Lineage evidence | Fastino own architecture; HF created 2026-08-18; prior art |
| DI 0.2.1 score | 3.82 (independent (DI 0.2.1 board)) |
| ECE (board) | 0.1608 |
| Median latency | 13.6 ms (DI board, 1x RTX PRO 6000, single process) |
| Hosted price (input) | self-hosted/M |
| Notes | Board raised a capacity limit ("lifted"). |
| Source | https://huggingface.co/fastino/gliner2.5-small-v1 |
All variants
10 variants
Lineage
finetune: 3quantized: 4adapter: 3
Top variants (10 shown, ranked by downloads)
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| litert-community/GLiNER2.5-Small-LiteRT | finetune | — | — | — | apache-2.0 | litert-community | 2 | 702 |
| FluidInference/gliner2-5-small-coreml | quantized | — | — | — | apache-2.0 | FluidInference | 0 | 80 |
| nicolasembleton/gliner2.5-small-v1-onnx | quantized | — | — | — | apache-2.0 | nicolasembleton | 1 | 0 |
| nativ-community/gliner2.5-small-v1 | finetune | — | 71.7M | S | apache-2.0 | nativ-community | 0 | 48 |
| pooria/foxmind-small | quantized | — | — | — | apache-2.0 | pooria | 0 | 31 |
| baobabtech/evalexplorer-classify-gliner2.5-small | finetune | — | 73.9M | S | apache-2.0 | baobabtech | 0 | 20 |
| na399/clinical-evidence-gliner2.5-small-lora-v3-act-sol | adapter | — | — | — | apache-2.0 | na399 | 0 | 18 |
| StephanAkkerman/stock-recognizer-gliner2.5-small | adapter | — | — | — | mit | StephanAkkerman | 0 | 14 |
| StephanAkkerman/options-recognizer-gliner2.5-small | adapter | — | — | — | mit | StephanAkkerman | 0 | 0 |
| EmogheneAkpomedaye/gliner2.5-small-v1-onnx | quantized | — | — | — | apache-2.0 | EmogheneAkpomedaye | 0 | 0 |
Repository changelog
20 commits
Repository changelog 20 commits
| Date | Commit | SHA |
|---|---|---|
| 2026-10-08 | Load this checkpoint as a Transformers schema-extraction model. (#9) | df5910e |
| 2026-09-28 | Point the fine-tune banner to agent.fastino.ai | 7132dc4 |
| 2026-09-24 | Update the Fastino GLiNER skill | 3ec6d3d |
| 2026-09-24 | Update GLiNER2 deployment badge link (#8) | bf8f2e7 |
| 2026-09-24 | Update GLiNER2.5 fine-tuning badge link (#7) | 1d23e38 |
| 2026-09-24 | Update Fastino hosted workflow links (#6) | d50ae4e |
| 2026-09-24 | Update the Fastino GLiNER skill | b0b7db2 |
| 2026-09-24 | Mention hosted fine-tuning via the Fastino API (#5) | e61ace6 |
| 2026-09-24 | Update the Fastino GLiNER skill | 98fe059 |
| 2026-09-24 | Add the Fastino GLiNER skill | c696d5f |
| 2026-09-20 | update button description (#4) | 7e6f537 |
| 2026-09-17 | upload new banner (#2) | f1e4d8f |
| 2026-09-17 | update banner and URL (#3) | d082704 |
| 2026-08-22 | Update README.md | cab1bdd |
| 2026-08-21 | Upload folder using huggingface_hub | d7e7d77 |
| 2026-08-20 | Add example outputs to every usage snippet. | 30470f1 |
| 2026-08-20 | Add Fastino model card banner. | 309d1df |
| 2026-08-20 | Add GLiNER2.5 boundary architecture model card and banner. | 8c39374 |
| 2026-08-18 | Upload folder using huggingface_hub | 5693482 |
| 2026-08-18 | initial commit | 5bebbe0 |
Papers
1 paper
Papers 1
arXiv links found in the model card and its tags. "Peer-reviewed" is shown only with evidence: a published venue in OpenAlex (measured) or the authors' own journal reference on arXiv (reported).
Similar models
3 models
Similar models
- gliner2.5-base-v1 · 194M same size band (S), same task (None), permissive license
- system-one-mini · 69.3M same size band (S), permissive license
- gliner2.5-multi-v1 · 287M same size band (S), same task (None), permissive license
Sources
3 sources
Sources
Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:44 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:46 UTC · Summary: 2026-10-07 15:19 UTC · Last verified 2026-10-25
Follow gliner2.5-small-v1
Watch this model
- /feeds/models/fastino--gliner2.5-small-v1.xml(RSS 2.0) · every change to gliner2.5-small-v1
- /feeds/families/fastino--gliner2.5-small-v1.xml(RSS 2.0) · every change across the gliner2.5-small-v1 family
Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.