Model passport
Oasst Pythia 12b
oasst-pythia-12b- Open weights
- Apache 2.0 licence
- Released 2023-04-15(curated)
- Overall · LMArena
- 1,024#391 of 396
- Best percentile
- 2.0Coding
- Cohort standing
- 22.5median 24.9
What the data says
Benchmarked on 4 of 27 tracked axes. Every figure below is the model's own published result on that board.
Weakest axes
- 1.5 pct — Overall (1,024)
- 2 pct — Hard prompts (1,023)
- 2 pct — Instruction following (988)
Uneven: 0.5 percentile points between its best and worst axis — 2 pct on Coding versus 1.5 pct on Overall.
- Coding2 pct1,050Verify →
- Instruction following2 pct988Verify →
- Hard prompts2 pct1,023Verify →
- Overall1.5 pct1,024Verify →
- WebDev ArenaNot benchmarked
- Vision ArenaNot benchmarked
- Factuality ArenaNot benchmarked
- Search ArenaNot benchmarked
- Document ArenaNot benchmarked
- Agent ArenaNot benchmarked
- GPQA DiamondNot benchmarked
- AIME (math)Not benchmarked
- FrontierMathNot benchmarked
- SimpleQA VerifiedNot benchmarked
- ARC-AGI-2Not benchmarked
- Humanity's Last ExamNot benchmarked
- SimpleBenchNot benchmarked
- LiveBenchNot benchmarked
- BFCL v4 (function calling)Not benchmarked
- SWE-bench Verified (live board)Not benchmarked
- Terminal-Bench 4.0Not benchmarked
- MMLU-Pro (Artificial Analysis)Not benchmarked
- LiveCodeBench (Artificial Analysis)Not benchmarked
- IFBench (Artificial Analysis)Not benchmarked
- HHEM Hallucination (Vectara)Not benchmarked
- SWE-bench VerifiedNot benchmarked
- Terminal-BenchNot benchmarked
Percentiles are the model's standing within each board's own population. Axes marked "not benchmarked" have no published result for this model — that is an absence of data, not a low score.
oasst-pythia-12b stands at 22.5 (raw mean 1.9) — measured on 4 of 27 axes.
oasst-pythia-12b stands at 22.5 versus a cohort median of 24.9 across 30 models released within 180 days — below the peer median.
oasst-pythia-12b scores at percentile 1.5 on Overall.
oasst-pythia-12b scores in the 2nd percentile on Coding.
oasst-pythia-12b ranks between #406 and #879 across 5 built-in weightings — best under "Conversation-first", worst under "All sources".
Every claim is a computed fact from the current board data, phrased by a low-temperature model call and rejected by a validation gate if any number is not in the source row. How the leaderboards work →