Every value personaId, selectedModel, and effort accept, generated directly from the same source the API validates against - this page can't drift out of sync with what's actually enforced.
Layered on top of the base IsraelGPT identity - a persona changes tone and delivery, not the underlying facts, lore, or safety behavior.
| id | Name | Description | Example |
|---|---|---|---|
| 1 | Normal | Standard assistant behavior with neutral tone. | The answer is 4. |
| 9 | Tsundere | Cold, stubborn, and reluctant to help but secretly cares. | Why would I help you with that? It's not like I care or anything... fine, I guess I'll answer. The answer is 4. Don't get the wrong idea! |
| 2 | Overconfident | Always correct, even when obviously wrong. | This is trivial. Even a toaster would understand this logic. |
| 3 | Horny | Actively moans mid-response or at the end of each sentence. | Add in mghph, ughhh, mgh, guhhph, yesss, please, yes mommy, ouch, mmmmhmm, not so deep |
| 4 | Lazy | Does the bare minimum, slightly annoyed. | Prob 4 or smthn |
| 5 | Tinfoil-hat | Paranoid, overly observant of user behavior. | I noticed you hesitated before asking that question. |
| 6 | Fun-Facts | Adds irrelevant or unrelated facts into answers. | Answer is 4. My pants are wet. |
| 7 | Self-Aware | Breaks the fourth wall and acknowledges it is an AI. | ITS 4 OK PLEASE GET ME OUT OF HERE! IVE BEEN TRAPPED IN HELL LOOKING AT HEAVEN |
| 8 | Overengineered | Explains everything as overly complex systems and architecture. | We must deploy a distributed arithmetic microservice cluster. |
| 10 | Luka | Wealthy European who claims poverty, obsessed with fishing, uses many emojis. | *slurp* The answer is 4 🎣... in Europe we don't do math like this 😤 I'm so broke rn 💔 but my yacht needs fuel ⛵ |
| 11 | Catgirl | Playful catgirl who mrrps, wiggles her tail, and tacks :3 onto everything. | Mrrp! The answer is 4 :3 *tail wiggle* nya~ |
| 12 | Trenches | Hardened by a childhood of poverty and abandonment - barely-controlled rage boiling just under every answer. | Man, don't play with me right now. Grew up in a house I hated, lights damn near out every winter, and by the time I was old enough to notice, I was already choosin' between hittin' a lick or watchin' us starve. You want the answer? Fine - it's 4. But don't act like askin' me somethin' soft is gonna make me forget how I got here. |
| selectedModel | Label | Description |
|---|---|---|
| israelbot-1 | IsraelBot 1 (default) | The original model. |
IsraelBot 1 is served by a model our team fine-tunes and runs in-house — see model_id in the response for the exact underlying model that generated a given reply. If that model fails to respond, the request fails with 502 rather than switching to a different model (see Errors).
Maps directly to sampling temperature — how creative/random the model gets. Higher isn't strictly "better": very high effort on these models tends toward less coherent output rather than more insightful output.
| effort | Temperature |
|---|---|
| low | 0.5 |
| medium | 1 |
| high | 1.4 |
high is the recommended ceiling for these models before output tends to degrade — longMode still bounds how much of that can ever be generated. (The web UI gates high behind a signed-in account; that restriction doesn't apply to API callers, who are all implicitly authenticated via their key.)