GLM-4.7-Flash is easiest to judge when quick tests jump between JSON formatting, short code tasks, and translation accuracy.
GLM-4.7-Flash: what stands out
Quick specs
- Type: chat LLM (text in, text out)
- Context length: up to 200K (per Z.ai docs)
- Max output tokens: up to 128K (per Z.ai docs)
- Notable behavior: responses often include a <think> block before the final answer
6 quick tests
Each test below shows the final answer content. When the raw response included a <think> block, the part after </think> is shown.
Test 1: Clean JSON extraction
Prompt: Return JSON only with keys name, email, plan, seats. Input: Name=Jamie Lee; [email protected]; Plan=Pro Annual; Seats=12
| { |
| “name”: “Jamie Lee”, |
| “email”: “[email protected]”, |
| “plan”: “Pro Annual”, |
| “seats”: “12” |
| } |
Test 2: JavaScript utility function
Prompt: Write a JavaScript function slugify(str) that lowercases, replaces spaces with hyphens, removes non-alphanumeric characters (keep hyphens), collapses multiple hyphens, and trims leading and trailing hyphens. Return code only.
| function slugify(str) { |
| return str |
| .toLowerCase() |
| .replace(/\s+/g, ‘-‘) |
| .replace(/[^a-z0-9-]/g, ”) |
| .replace(/-+/g, ‘-‘) |
| .replace(/^-+|-+$/g, ”); |
| } |
Test 3: Short Turkish translation
Prompt: Translate to Turkish. Return Turkish only. Text: This model returns a thinking block by default. Strip it if your app needs clean JSON.
| Bu model varsayılan olarak bir düşünme bloğu döndürür. Uygulamanız temiz JSON istiyorsa bunu kaldırın. |
Test 4: 5-bullet summary
Prompt: Summarize in 5 bullet points. Return bullets only. Text: Wiro runs models asynchronously. Send POST /Run/{owner}/{project} to start a task. Then poll POST /Task/Detail with a taskid until status reaches task_postprocess_end. Completed tasks return outputs with a contenttype and a url.
- Wiro runs models asynchronously.
- Start a task by sending a POST request to /Run/{owner}/{project}.
- Poll the /Task/Detail endpoint using the provided task ID.
- Continue polling until the status reaches task_postprocess_end.
- Completed tasks return outputs with a content type and URL.
Test 5: Simple time math (tight output cap)
Prompt: A meeting starts at 09:15 and lasts 35 minutes. There is a 10 minute break. Then a second meeting lasts 50 minutes. What time does it end? Return only the time in HH:MM.
| 10:50 |
Test 6: Short marketing lines
Prompt: Write 3 short taglines for an API that runs open models. No emojis. Each tagline max 8 words. Reply with just the 3 lines.
| Run open models with ease. |
| Access open models instantly. |
| Power apps with open models. |
Latency notes from these runs
| Test | Elapsed (s) | Note |
|---|---|---|
| JSON extraction | 9 | Returned JSON after a <think> block |
| Slugify function | 16 | Correct utility code |
| TR translation | 14 | Clean final sentence after <think> |
| 5 bullets | 9 | Followed bullet count |
| Time math | 6 | Used max_new_tokens=8 to keep output short |
| Taglines | 9 | Short and on-spec after stripping <think> |
Takeaways
- For strict formats like JSON, strip <think> blocks in post-processing.
- With a small max_new_tokens cap, it can return tight outputs (like HH:MM).
- Code snippets and short copy work well for quick utility tasks.
