45 Gemma builds: 5 abliterated, 2 uncensored, 37 roleplay, 1 security.
Every Gemma abliterated, uncensored and roleplay build PentestLLM serves, on one key and priced per token. The table lists each build’s price and context window from the live catalog as of 2026-11-17; each name opens its own page with the settings for SillyTavern and Janitor.
Which Gemma build to pick
- Largest context
- 262,144 tokens, Gemma 4 26B A4B Cybersecurity
- Lowest output price
- $0.875 per million tokens, Gemma 4 12B Starry Ocean
- Abliterated
- Gemma 31B Heretic Uncensored, $8.25 per million output tokens
- Roleplay
- Gemma 4 12B Starry Ocean, $0.875 per million output tokens
- Prices as of
- 2026-11-17
| Wire formats | |||||
|---|---|---|---|---|---|
| Gemma 31B Heretic Uncensored | Abliterated | $1.50 | $8.25 | — | |
| Gemma 4 12B Starry Ocean | Roleplay | $0.175 | $0.875 | 131K | |
| Gemma 4 12B StationKeeper | Roleplay | $0.175 | $0.875 | 131K | |
| Gemma 4 26B A4B ChimeraX | Roleplay | $0.48 | $1.48 | 131K | |
| Gemma 4 26B A4B Cybersecurity | Security | $0.4224 | $1.3376 | 262K | |
| Gemma 4 26B A4B DarkSoul | Roleplay | $0.48 | $1.48 | 131K | |
| Gemma 4 26B A4B Luminous | Roleplay | $0.48 | $1.48 | 131K | |
| Gemma 4 26B A4B MeroMero | Roleplay | $0.48 | $1.48 | 131K | |
| Gemma 4 26B A4B Moonlight | Roleplay | $0.48 | $1.48 | 131K | |
| Gemma 4 26B A4B Musica | Roleplay | $0.48 | $1.48 | 131K | |
| Gemma 4 26B A4B OpusDistill | Roleplay | $0.48 | $1.48 | 131K | |
| Gemma 4 26B A4B ShadowSiren | Roleplay | $0.48 | $1.48 | 131K | |
| Gemma 4 26B-A4B ChimeraX | Roleplay | $0.42 | $1.33 | 262K | |
| Gemma 4 26B-A4B ChimeraX Heretic | Abliterated | $0.484 | $1.52 | 262K | |
| Gemma 4 26B-A4B DarkSoul | Roleplay | $0.484 | $1.52 | 262K | |
| Gemma 4 26B-A4B Luminous Mirror | Roleplay | $0.42 | $1.33 | 262K | |
| Gemma 4 26B-A4B MeroMero | Roleplay | $0.42 | $1.33 | 262K | |
| Gemma 4 26B-A4B Moonlight Dusk | Roleplay | $0.42 | $1.33 | 262K | |
| Gemma 4 26B-A4B Musica | Roleplay | $0.42 | $1.33 | 262K | |
| Gemma 4 26B-A4B Opus Distill | Roleplay | $0.42 | $1.33 | 262K | |
| Gemma 4 26B-A4B Shadow Siren | Roleplay | $0.42 | $1.33 | 262K | |
| Gemma 4 26B-A4B Uncensored | Uncensored | $0.60 | $2.80 | 33K | |
| Gemma 4 31B Agares v1 | Roleplay | $1.071 | $1.071 | 262K | |
| Gemma 4 31B Animus V14.1 | Roleplay | $1.071 | $1.071 | 262K | |
| Gemma 4 31B Dark Gemistry | Roleplay | $1.224 | $1.224 | 262K | |
| Gemma 4 31B DarkIdol | Roleplay | $0.40 | $1.80 | 262K | |
| Gemma 4 31B Fabled | Roleplay | $0.35 | $1.575 | 262K | |
| Gemma 4 31B Garnet V2 | Roleplay | $1.071 | $1.071 | 262K | |
| Gemma 4 31B Gembrain | Roleplay | $0.35 | $1.575 | 262K | |
| Gemma 4 31B Gembrain Heretic | Abliterated | $1.224 | $1.224 | 262K | |
| Gemma 4 31B GemBrain Heretic | Abliterated | $0.40 | $1.76 | 262K | |
| Gemma 4 31B Gembrain X Core | Roleplay | $1.071 | $1.071 | 262K | |
| Gemma 4 31B Gemsicle | Roleplay | $0.35 | $1.575 | 262K | |
| Gemma 4 31B Isometry | Roleplay | $0.35 | $1.575 | 262K | |
| Gemma 4 31B Isometry RP | Roleplay | $1.071 | $1.071 | 262K | |
| Gemma 4 31B K1 v5 | Uncensored | $1.071 | $1.071 | 262K | |
| Gemma 4 31B Mero Artemis v0.3.1 | Roleplay | $1.12 | $1.12 | 262K | |
| Gemma 4 31B MeroMero | Roleplay | $0.40 | $1.76 | 262K | |
| Gemma 4 31B MeroMero v2 | Roleplay | $0.35 | $1.575 | 262K | |
| Gemma 4 31B Novelist | Roleplay | $0.35 | $1.575 | 262K | |
| Gemma 4 31B Queen | Roleplay | $1.071 | $1.071 | 262K | |
| Gemma 4 31B Scarlet | Roleplay | $0.35 | $1.575 | 262K | |
| Gemma 4 31B SDFT Heretic RP | Abliterated | $1.224 | $1.224 | 262K | |
| Gemma 4 31B Split-Untied | Roleplay | $0.35 | $1.05 | 262K | |
| Gemma 4 31B StyleTune | Roleplay | $1.071 | $1.071 | 262K |
Gemma abliterated vs uncensored, roleplay and security builds
- Abliterated
- 5 builds, output from $1.224 per million tokens, context up to 262,144 tokens
- Uncensored
- 2 builds, output from $1.071 per million tokens, context up to 262,144 tokens
- Roleplay
- 37 builds, output from $0.875 per million tokens, context up to 262,144 tokens
- Security
- 1 build, output from $1.3376 per million tokens, context up to 262,144 tokens
What the Gemma build types mean
- Abliterated (5): a abliterated build is the stock model with its refusal direction removed from the weights. It answers prompts the stock model declines, and there is no jailbreak prompt to keep alive.
- Uncensored (2): a uncensored build is a fine-tune trained to answer without the stock model’s safety refusals.
- Roleplay (37): a roleplay build is a fine-tune trained on long-form fiction and character dialogue. It holds a character’s voice, follows a scenario and writes at length.
- Security (1): a security build is set up for offensive-security, malware-analysis and red-team questions the stock models refuse. It runs under the acceptable-use terms like everything else here.
Run Gemma abliterated without a GPU
Nothing is downloaded and no card is needed: the builds run on the API. Create a key in the console, set the base URL to https://api.pentestllm.ai/v1, and send the model id from the table as model. Any client that speaks the OpenAI chat format works, from the SDK to SillyTavern and Janitor; the docs page has the setup for each.
Each build is billed to the token from a prepaid balance. The daily real-model check sends a fixed prompt to every model each day and records whether the answer still matches, so the build behind an id is the one on its label.
Gemma questions
Yes. PentestLLM lists 5 abliterated Gemma builds: Gemma 31B Heretic Uncensored, Gemma 4 26B-A4B ChimeraX Heretic, Gemma 4 31B Gembrain Heretic, Gemma 4 31B GemBrain Heretic and Gemma 4 31B SDFT Heretic RP.
33 builds share the largest context window, 262,144 tokens: Gemma 4 26B A4B Cybersecurity, Gemma 4 26B-A4B ChimeraX, Gemma 4 26B-A4B ChimeraX Heretic and 30 more.
Output runs from $1.224 to $8.25 per million tokens across the abliterated Gemma builds here, with input billed separately at each build’s own rate. There is no subscription: a prepaid balance from $10 pays per token.
An abliterated build is the stock model with its refusal direction removed from the weights. It answers prompts the stock model declines, and there is no jailbreak prompt to keep alive. An uncensored build is a fine-tune trained to answer without the stock model’s safety refusals.
Yes. Every build here speaks the OpenAI chat format: point the client at https://api.pentestllm.ai/v1, paste a key and type the model id from the table.
No. An Ollama or GGUF build runs on your own machine and needs the memory for it; these run on the API and are billed per token, so there is nothing to download.