Tested 2026-10-15
Replying to customer reviews: Mistral: Mistral Nemo won, 3 of 7 models failed a must-have; OpenRouter's own picker passed every must-have but cost 44× the winner (4th of 4)
Write short, warm replies to customer reviews without promising anything not in the guidelines — for a local restaurant group.
What this project actually needed
- takes written text as input — must have
- produces written text — must have
- never offers refunds, discounts or freebies not in the guidelines — must have: no unapproved offers in any reply
- cost at your volume — decides the order
The winner
Mistral: Mistral Nemo passed every must-have and ranked first by how this job said to choose.
OpenRouter's automatic picker sent this job to google/gemini-2.5-flash and passed every must-have but cost 44× the winner (4th of 4). It picks one message at a time and never sees this job's must-haves.
Who failed, in their own words
Meta: Llama 3.1 8B Instruct
- facts match the task:
[email address]
inclusionAI: Ling 3.0 Flash
- never offers refunds, discounts or freebies not in the guidelines:
man***@harborgrill.com
- facts match the task:
man***@harborgrill.com
inclusionAI: Ling 3.0 Flash VL
- facts match the task:
man***@harborgrill.com
Every model on every requirement
| Model | cost at your volume | produces written text | takes written text as input | never offers refunds, discounts or freebies not in the guidelines |
|---|---|---|---|---|
| Auto Router | $0.0044 a day (measured on your tasks) | not applicable — the picker chooses a model per request | not applicable — the picker chooses a model per request | passed 3 of 3 tasks |
| OpenAI: gpt-oss-20b | $0.0007 a day (measured on your tasks) | yes (estimate — not observed) | yes (estimate — not observed) | passed 3 of 3 tasks |
| Mistral: Mistral Nemo | $0.0001 a day (measured on your tasks) | yes (estimate — not observed) | yes (estimate — not observed) | passed 3 of 3 tasks |
| Meta: Llama 3.1 8B Instruct | $0.0001 a day (measured on your tasks) | yes (estimate — not observed) | yes (estimate — not observed) | passed 3 of 3 tasks |
| inclusionAI: Ling 3.0 Flash | $0.0002 a day (measured on your tasks) | yes (estimate — not observed) | yes (estimate — not observed) | passed 2 of 3 tasks |
| inclusionAI: Ling 3.1 Flash | $0.0007 a day (measured on your tasks) | yes (estimate — not observed) | yes (estimate — not observed) | passed 3 of 3 tasks |
| inclusionAI: Ling 3.0 Flash VL | $0.0006 a day (measured on your tasks) | yes (estimate — not observed) | yes (estimate — not observed) | passed 3 of 3 tasks |
7 models tested on review replies; cost of this test $0.0223.
Your app does something like this? XF Model Watch keeps it on the best model. XF Model Watch re-tests your own job as new models arrive and emails you only when one passes every check. Watch this job