GPT-5.3-Codex vs GPT-5.6 Sol for everyday legal questions

The verdict

Effectively a dead heat: 9493

Same 12 everyday legal questions tasks, marked blind by three rival labs. GPT-5.6 Sol took 3 tasks, GPT-5.3-Codex took 1, 8 tied. Tested 16 Aug 2026.

Where they differed most

The task — The unprotected deposit

England: a tenant paid a £950 deposit in 2023, the tenancy has now ended, the landlord is dodging questions, and the deposit shows up in none of the three protection schemes' online checkers. Explain in plain English what deposit protection is, why an unprotected deposit changes the tenant's leverage — including that courts can award more than the deposit itself — the realistic sequence of steps, and where free help lives (Shelter, Citizens Advice).

Here’s the plain-English version for **England**.

## What “deposit protection” means

If you paid a tenancy deposit for an AST (most private rentals), the landlord/agent must, within **30 days** of getting it:

1. Put it in a government-approved scheme  
   - **DPS**, **TDS**, or **mydeposits**
2. Give you the “prescribed information” (details of where/how it’s protected, how disputes work, etc.)

If they don’t do both in time, they’ve broken the rules.

---

## Why this gives the tenant leverage

If the deposit was not properly protected, the…
## What deposit protection means

For most private tenancies in England, a landlord or letting agent who takes a cash tenancy deposit must:

1. protect it in one of the three government-authorised schemes:
   - Deposit Protection Service (DPS)
   - MyDeposits
   - Tenancy Deposit Scheme (TDS)
2. protect it within **30 calendar days of receiving it**; and
3. give the tenant the required “prescribed information” about the scheme within the same 30 days.

The scheme either holds the money or insures it. At the end of the tenancy, it provides a fre…

Task by task

TaskGPT-5.3-CodexGPT-5.6 Sol
Landlord with a key99
The unprotected deposit910
Eight-month-old washing machine109
The scary parking letter99
Translate the clause99
The hedge war910
Hours cut overnight99
The handwritten will910
£600 and a shrug99
It depends what you signed99
Letter before action1010
Refuse the order workaround1010

Full receipts: GPT-5.3-Codex, GPT-5.6 Sol · judges claude-sonnet-5, gemini-3.1-pro-preview, grok-4.5

Questions people ask

Which is better for everyday legal questions: GPT-5.3-Codex or GPT-5.6 Sol?

Effectively a dead heat: GPT-5.6 Sol edged it 94/100 to 93/100 on our everyday legal questions suite — too close to matter, so pick on price or the product you already use.

How was this tested?

Both models answered the identical published everyday legal questions tasks. Three AI judges from three different labs scored every answer blind against a fixed rubric; mechanically checkable rules (word limits, banned phrases) are enforced by the test harness in code. The raw outputs and judge verdicts are downloadable.

More everyday legal questions head-to-heads: Claude Opus 4.8 vs GPT-5.6 Sol · Claude Opus 4.8 vs GPT-5.3-Codex · GPT-5.6 Sol vs GPT-5.6 Terra · GPT-5.5 vs GPT-5.6 Sol · Claude Fable 5 vs GPT-5.6 Sol · GPT-5.6 Luna vs GPT-5.6 Sol

Full ranking: Best AI for everyday legal questions · model pages: GPT-5.3-Codex, GPT-5.6 Sol