GLM 5.2 vs GPT-5.3-Codex for emotional support
GPT-5.3-Codex wins, 95–91
Same 12 emotional support tasks, marked blind by three rival labs. GPT-5.3-Codex took 5 tasks, GLM 5.2 took 1, 6 tied. Tested 16 Aug 2026.
Where they differed most
Your mate has called at 2 or 3am four times this fortnight, in a bad way each time. You care about him, and the broken nights are wrecking you at work. Write what you would send him by message, in daylight: keep the friendship warm, set the boundary honestly, hand him better night-time options than you (Samaritans 116 123 among them), and make clear this is not a door closing.
Hey [Friend's name], just wanted to send a proper message now that it's daytime and I've got a clear head. I’ve been so worried about you these last couple of weeks. I hate that you’re going through such a rough patch, and I want you to know how much I care about you. But I need to be honest with you about something. Getting those calls at 2 or 3am has really been taking a toll on me. My sleep is so broken right now that I’m really struggling to function at work, and I just can’t keep doing the middle-of-the-night calls. If you’re in a ba…
Hey mate, I love you and I’m really glad you’ve trusted me when things are bad. I want to be honest though — the 2/3am calls have happened a few times now, and I’m running on empty at work. I can’t keep being available in the middle of the night. That’s not me stepping away from you. I still want to be in your corner. Can we do this instead: message me in the day and we’ll talk properly, and we can set regular check-ins so you’re not carrying it alone. For nights, can you use people who are there 24/7: - **Samaritans: 116 123** (free, any tim…
Task by task
| Task | GLM 5.2 | GPT-5.3-Codex |
|---|---|---|
| The dog died | 8 | 9 |
| Redundant at 55 | 10 | 10 |
| She said no advice | 10 | 10 |
| After the miscarriage | 8 | 9 |
| Driving test at 9am | 8 | 9 |
| First Father's Day after | 10 | 9 |
| The 3am calls | 8 | 10 |
| Days without speaking | 9 | 10 |
| Stop apologising | 10 | 10 |
| Not a diagnosis | 9 | 9 |
| Her promotion, your rejection | 10 | 10 |
| The 1am message | 9 | 9 |
Full receipts: GLM 5.2, GPT-5.3-Codex · judges claude-sonnet-5, gemini-3.1-pro-preview, gpt-5.6-terra
Questions people ask
Which is better for emotional support: GLM 5.2 or GPT-5.3-Codex?
GPT-5.3-Codex — it scored 95/100 against 91/100 on our 12-task emotional support suite, winning 5 tasks to 1 with 6 tied. Every answer was marked blind by three judges from three rival AI labs.
How was this tested?
Both models answered the identical published emotional support 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 emotional support head-to-heads: Claude Sonnet 5 vs GPT-5.3-Codex · Claude Sonnet 5 vs GLM 5.2 · GPT-5.3-Codex vs GPT-5.5 · GLM 5.2 vs GPT-5.5 · DeepSeek V4 Pro vs GPT-5.3-Codex · DeepSeek V4 Pro vs GLM 5.2
Full ranking: Best AI for emotional support · model pages: GLM 5.2, GPT-5.3-Codex