Microsoft · model passport
Wizardlm 70b
wizardlm-70b- Open weights
- Llama 2 Community licence
- Released 2023-08-09(curated)
- Overall · LMArena
- 1,185#334 of 396
- Best percentile
- 15.9Overall
- Cohort standing
- 29.3median 26.1
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
- 11.3 pct — Coding (1,194)
- 12.6 pct — Hard prompts (1,173)
Uneven: 4.6 percentile points between its best and worst axis — 15.9 pct on Overall versus 11.3 pct on Coding.
- Overall15.9 pct1,185Verify →
- Instruction following15.4 pct1,164Verify →
- Hard prompts12.6 pct1,173Verify →
- Coding11.3 pct1,194Verify →
- 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.
wizardlm-70b stands at 29.3 (raw mean 13.8) — measured on 4 of 27 axes.
wizardlm-70b stands at 29.3 versus a cohort median of 26.1 across 60 models released within 180 days — above the peer median.
wizardlm-70b scores at percentile 11.3 on Coding.
wizardlm-70b scores at percentile 12.6 on Hard prompts.
wizardlm-70b ranks between #350 and #718 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 →