gliclass:large
GLiClass · Knowledgator · 439M · ONNX Runtime · Apache-2.0
- Score
- 15.3
- 95 % interval 13.1–17.6
- Rank
- 31
- of 35 scored models
- Public benchmark
- 43.4
- macro accuracy, %
- typed-decisions
- 47.7
- accuracy, %
By task
Chance-corrected accuracy on each of the 14 tasks: 0 is guessing, 100 is every answer right.
Accuracy and calibration
| Benchmark | Accuracy | Choice | Noul | Score | ECE | Brier | Rejected |
|---|---|---|---|---|---|---|---|
| Public benchmark | 43.4(41.9–44.8) | 36.6 | 60.0 | 26.9 | 0.284 | 0.810 | 28 / 3,880 |
| typed-decisions | 47.7(45.3–50.2) | 42.7 | 62.7 | 40.3 | 0.219 | 0.742 | 0 / 400 requests |
Accuracy in %, with its 95 % bootstrap interval. A rejected request (the model can’t take the question, or the input is too long for it) counts as wrong.
Speed and memory
| Device | p50 | p90 | Load | VRAM | RAM peak | Precision |
|---|---|---|---|---|---|---|
| RTX 5090 CUDA | 14.1 ms | 14.5 ms | 7.15 s | 2.6 GB | 2.8 GB | F32 |
| Threadripper 3960X CPU | 2.13 s | 2.19 s | 6.79 s | – | 2.4 GB | F32 |
Five-question requests over HTTP, one at a time, each a different short message. VRAM: the rise of the GPU’s used memory while the model is loaded. RAM peak: the runner process’s peak resident memory.
Parity with the authors’ code
Its parity check is in the model’s family documentation in the repository (docs/families/); the compiled parity file does not list it yet.
Reproduce
The tools are in convert/ollaya_convert and need only Python’s standard library. Run them against a local Ollaya 0.13.0 with the same model digest.
ollaya pull gliclass:large
# typed-decisions; export the test split once with
# uv run python -m ollaya_convert.bench_typed export --out typed.jsonl
python -m ollaya_convert.bench_typed run --model gliclass:large \
--url http://127.0.0.1:11435 --data typed.jsonl --out typed/gliclass-large
# Bespoke Labs' public benchmark; build its 13 subsets first
NIMBLE_SRC=nimble python -m ollaya_convert.bench_public run --model gliclass:large \
--url http://127.0.0.1:11435 --data public --out public/gliclass-large
# speed and memory
python -m ollaya_convert.bench_latency --models gliclass:large --memory \
--machine my-machine --device cuda --out latency.json