xAI · model passport
Grok 4.0709
grok-4-0709- Cloud only
- Proprietary licence
- Released 2025-07-09(curated)
- Overall · LMArena
- 1,411#136 of 396
- Best percentile
- 91.5BFCL v4 (function calling)
- Cohort standing
- 56.8median 51.5
What the data says
Benchmarked on 14 of 27 tracked axes. Every figure below is the model's own published result on that board.
Strongest axes
- 91.5 pct — BFCL v4 (function calling) (63.0%)
Uneven: 66.5 percentile points between its best and worst axis — 91.5 pct on BFCL v4 (function calling) versus 25 pct on Terminal-Bench.
- BFCL v4 (function calling)Dormant board91.5 pct63.0%Verify →
- GPQA Diamond72.4 pct87.0%Verify →
- SimpleBench70.1 pct60.5%Verify →
- LiveBench69.8 pct64.0%Verify →
- Overall65.9 pct1,411Verify →
- Instruction following65.2 pct1,396Verify →
- Hard prompts63.9 pct1,420Verify →
- AIME (math)60 pct84.0%Verify →
- Coding59.1 pct1,435Verify →
- FrontierMath56.5 pct19.7%Verify →
- ARC-AGI-248.3 pct16.0%Verify →
- Vision Arena40.7 pct1,184Verify →
- Factuality Arena27.9 pct1,418Verify →
- Terminal-Bench25 pct27.2%Verify →
- WebDev ArenaNot benchmarked
- Search ArenaNot benchmarked
- Document ArenaNot benchmarked
- Agent ArenaNot benchmarked
- SimpleQA VerifiedNot benchmarked
- Humanity's Last ExamNot 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
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.
grok-4-0709 stands at 56.8 (raw mean 58.3) — measured on 14 of 27 axes.
grok-4-0709 stands at 56.8 versus a cohort median of 51.5 across 145 models released within 180 days — above the peer median.
grok-4-0709 scores at percentile 91.5 on BFCL v4 (function calling).
On GPQA Diamond, grok-4-0709 raised the running best score by 1.7 percentage points to 87% on 2025-07-09.
On ARC-AGI-2, grok-4-0709 raised the running best score by 7.4 percentage points to 16% on 2025-07-09.
grok-4-0709 ranks between #95 and #276 across 6 built-in weightings — best under "All-rounder", worst under "Coding-focused".
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 →