Head-to-head · published results only

GPT-5.6 Luna vs Grok 4.6 on finance benchmarks

The best-ranked OpenAI and SpaceXAI entries on Finance Agent v2. GPT-5.6 Luna is ahead on analyst research and Excel modeling at a fraction of the price; Grok 4.6 is ahead on corporate tax research.

Same benchmarks, same harness, as published by Vals AI Context: FAB v2: n=450 test questions, 3 runs · Oct 12, 2026 leaderboard · Vals AI · ± SEM Context: EMB: n=51 test tasks · Oct 12, 2026 leaderboard · Vals AI · ± as published, type not stated Context: Tax Agent: n=193 test questions · Oct 12, 2026 leaderboard · Vals AI · ± as published, type not stated
Measure	GPT-5.6 Luna	Grok 4.6	Gap	Read
Finance Agent v2	55.04% ±0.31	53.68% ±0.67	1.36 pts	GPT-5.6 Luna ahead
Excel Modeling Benchmark	67.12% ±2.94	62.73% ±3.08	4.39 pts	Within published margins
Tax Agent Bench	24.82% ±3.11	33.14% ±3.40	8.32 pts	Grok 4.6 ahead
FAB v2 cost / time per task	$0.28 / 12m52s	$1.66 / 16m41s
API price per 1M tokens (in / out)	$0.20 / $1.20	$2.00 / $6.00

How to read this

"Within published margins" means the gap is smaller than the two published margins added together, so the order is not settled.
Both models ran in the same Vals AI harness with the same tools; your own prompts and tools may give different results.
These are research and modeling tests graded by AI judges, not trading results.

Limits that apply to each model

GPT-5.6 Luna

On Tax Agent Bench it timed out on 8 of 193 test questions (4.1%) Context: n=193 test questions · Oct 12, 2026 leaderboard · Vals AI · not applicable (count), and each timeout scores zero.
GPT-5.4, an earlier OpenAI model, sits on the Finance Agent v2 jury.

Grok 4.6

Below the leaders on both analyst research and Excel modeling (30th on each).
Search-heavy research style on Tax Agent Bench, according to Vals, which does not by itself predict accuracy.

Primary sources

Verified
GPT-5.6 Luna model page (Vals AI)Checked October 14, 2026.
Verified
Grok 4.6 model page (Vals AI)Checked October 14, 2026.
Verified
Finance Agent v2 (Vals AI)Checked October 14, 2026.
Verified
Excel Modeling Benchmark (Vals AI)Checked October 14, 2026.
Verified
Tax Agent Bench (Vals AI)Checked October 14, 2026.

Last verified: Oct 14, 2026. Prices and scores change; check the primary source before relying on them. We have no affiliate links and receive nothing from the companies listed. Every percentage carries its sample size, period and source (no naked percentages).

Not financial advice. For information and education only. Benchmark scores and vendor descriptions are not trading results and do not predict returns. Nothing here is a recommendation to buy or sell any asset or to subscribe to any product. See our disclaimer.