Model passport
Guanaco 33b
guanaco-33b- Open weights
- Non-commercial licence
- Released 2023-05-23(curated)
- Overall · LMArena
- 1,128#371 of 396
- Best percentile
- 6.6Overall
- Cohort standing
- 24.1median 26.0
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
- 3.8 pct — Instruction following (1,066)
- 3.8 pct — Coding (1,114)
- 4.5 pct — Hard prompts (1,106)
Uneven: 2.8 percentile points between its best and worst axis — 6.6 pct on Overall versus 3.8 pct on Instruction following.
- Overall6.6 pct1,128Verify →
- Hard prompts4.5 pct1,106Verify →
- Coding3.8 pct1,114Verify →
- Instruction following3.8 pct1,066Verify →
- 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.
guanaco-33b stands at 24.1 (raw mean 4.7) — measured on 4 of 27 axes.
guanaco-33b stands at 24.1 versus a cohort median of 26 across 40 models released within 180 days — below the peer median.
guanaco-33b scores at percentile 3.8 on Coding.
guanaco-33b scores at percentile 3.8 on Instruction following.
guanaco-33b ranks between #391 and #819 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 →