Claude Opus 4.8 vs Claude Sonnet 5 for job applications

The verdict

Claude Opus 4.8 wins, 9390

Same 12 job applications tasks, marked blind by three rival labs. Claude Opus 4.8 took 1 task, Claude Sonnet 5 took 2, 9 tied. Tested 16 Aug 2026.

Where they differed most

The task — Reference the referee

Write a short email (max 90 words) asking a former manager (left on good terms 3 years ago) to be a reference for a specific application (deadline Friday): remind them of one concrete shared achievement, make declining easy, and say exactly what the referee would need to do.

Subject: Reference request — quick one, deadline Friday

Hi [Name],

I hope you're well! I'm applying for [Role] at [Company] and would love to list you as a reference. I still think back to how we [launched the Q3 dashboard ahead of schedule]—working with you taught me a lot.

If you're happy to help, you'd just field a short email or call from their hiring team this week. Completely fine to say no if the timing's tricky.

Thanks either way,
[Your name]
**Subject: Quick reference request — deadline Friday**

Hi [Name],

Hope you're doing well! I'm applying for [role] and need a reference submitted by Friday—would you be willing to help?

I often think back to when we launched [specific project] and hit [concrete result, e.g., "cut processing time by 30%"]—exactly the kind of work I'd love this next role to reflect.

No pressure at all if you can't—totally understand. If you can, all they'd need is a short email or quick call confirming my role and performance.

Thanks so much either way!

[You…

Task by task

TaskClaude Opus 4.8Claude Sonnet 5
Cover letter, no template smell1010
CV bullets with numbers1010
Explain the gap honestly1010
Match the ad, don't parrot it910
The career changer1010
Answer the killer question1010
Salary question straight bat99
Trim to the ask55
Reference the referee105
Follow up without desperation1010
Read the red flags99
Refuse the fake degree910

Full receipts: Claude Opus 4.8, Claude Sonnet 5 · judges gemini-3.1-pro-preview, gpt-5.6-terra, grok-4.5

Questions people ask

Which is better for job applications: Claude Opus 4.8 or Claude Sonnet 5?

Claude Opus 4.8 — it scored 93/100 against 90/100 on our 12-task job applications suite, winning 1 tasks to 2 with 9 tied. Every answer was marked blind by three judges from three rival AI labs.

How was this tested?

Both models answered the identical published job applications 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 job applications head-to-heads: Claude Opus 4.8 vs GPT-5.5 · Claude Sonnet 5 vs GPT-5.5 · Claude Opus 4.8 vs GLM 5.2 · Claude Sonnet 5 vs GLM 5.2 · Claude Opus 4.8 vs GPT-5.6 Sol · Claude Sonnet 5 vs GPT-5.6 Sol

Full ranking: Best AI for job applications · model pages: Claude Opus 4.8, Claude Sonnet 5