Starter Pack
$9 $10
+$1 bonus credit
Valid for 30 days · All models · Unlocks Standard Preset
Open consolePay as you go in USD per 1 million tokens (MTok) with balance that never expires, or save extra with a manual 30-day credit pack ($9 for $10, $19 for $23, or $39 for $48). No forced subscriptions and no 5.5% top-up fee.
| Model | SillyTavern Model ID | Context | Input | Output (incl. thinking) | Cache Read | Cache Write | Best for |
|---|---|---|---|---|---|---|---|
| Claude Opus 5.5 | claude-opus-5.5 | 1M | $2.80 | $14.00 | $0.14 | $3.50 | 30% off official rates. #1 overall roleplay experience: sharpest scene logic, psychology, and setting adherence. |
| Claude Sonnet 5.5 | claude-sonnet-5.5 | 1M | $1.60 | $8.00 | $0.08 | $2.00 | 20% off official rates. Available if you prefer Sonnet, though we recommend Opus 5.5 or Gemini 4 Argon first. |
| Gemini 4 Argon | gemini-4-argon | 1M | $1.20 | $6.00 | $0.06 | No fee | 40% off official rates. Best everyday pick for most players: #1 prose style and massive world knowledge. |
| Gemini 3.8 Flash | gemini-3.8-flash | 1M | $0.49 | $2.45 | $0.049 | No fee | ~35% off official rates. Fast, lively Gemini prose for high-volume daily chatting and testing new cards. |
| DeepSeek V4 Pro | deepseek-v4-pro | 1M | $1.49 | $4.49 | $0.05 | No fee | Official DeepSeek API at one flat rate 24/7. Strong value for English and Chinese roleplay with our preset. |
All five models come from official sources (Claude via AWS/GCP, Gemini via Google AI Studio, DeepSeek via official API) and support a 1M context window. Claude and Gemini are priced 20%–40% below official list rates; DeepSeek V4 Pro uses a flat all-day rate roughly equal to official peak-hour pricing. Gemini and DeepSeek do not charge a separate cache-write fee (first-turn writes are billed at the normal input rate).
In roleplay, your bill comes mostly from input tokens because SillyTavern sends your chat history on every turn. Here is what a typical 50-reply evening at 32k context (~1.6M input tokens + 20k output tokens) costs with and without prompt caching:
| Model | 50 replies without caching | 50 replies with TavernAPI caching | Estimated savings |
|---|---|---|---|
| Claude Opus 5.5 | ~$4.76 | ~$1.03 | ~78% less |
| Claude Sonnet 5.5 | ~$2.72 | ~$0.59 | ~78% less |
| Gemini 4 Argon | ~$2.04 | ~$0.43 | ~79% less |
| Gemini 3.8 Flash | ~$0.83 | ~$0.21 | ~75% less |
| DeepSeek V4 Pro | ~$2.47 | ~$0.44 | ~82% less |
You never have to subscribe—regular top-up balance never expires. If you play regularly and want extra credit, you can buy a manual 30-day credit pack in the console. Packs work on all five models and never auto-renew.
$9 $10
+$1 bonus credit
Valid for 30 days · All models · Unlocks Standard Preset
Open console$19 $23
+$4 bonus credit
Valid for 30 days · All models · Includes Pro Preset while active
Open console$39 $48
+$9 bonus credit
Valid for 30 days · All models · Includes Pro Preset & permanently unlocks it ($25+ spend)
Open consoleOur server-locked SillyTavern presets come in two tiers. Once unlocked, download the preset JSON from the Presets tab in the console and import it into SillyTavern.
Unlocked permanently after your first top-up of $5 or more.
Unlock in any of three ways: buy it standalone for $4.99, use it for free while an active $19 or $39 30-day credit pack is in effect, or unlock it permanently once your cumulative spend reaches $25.
We only carry models that roleplay players actually enjoy reading. GPT models tend to summarize feelings, repeat robotic phrases ('a testament to', 'shivers down your spine'), and lecture the user mid-scene. Instead of listing 300 models you won't use, we put our effort into official 1M-context routing and tuned SillyTavern presets for Claude, Gemini, and DeepSeek.
Verify your email address (or sign in with Google or Discord) to get $0.30 in free trial credit. You can test Gemini 4 Argon (gemini-4-argon) and Gemini 3.8 Flash (gemini-3.8-flash) in SillyTavern right away.
Top up between $5 and $500 using USDT or USDC (via NOWPayments) or redeem a TavernAPI code (TAV-XXXX-XXXX-XXXX) with no platform top-up fee. Regular top-up balance never expires.
Want to stretch your budget without a recurring subscription? Pick up a manual 30-day credit pack: pay $9 for $10, $19 for $23 (Recommended), or $39 for $48. Usable across all five models for 30 days, with zero auto-renewal.
Your first top-up of $5+ unlocks the Standard preset permanently. The Pro preset is available standalone for $4.99, included for free during an active $19 or $39 credit pack, or unlocked permanently once your cumulative spend hits $25.
Regular pay-as-you-go balance added via crypto top-ups or redeem codes never expires. If you want extra credit without a recurring subscription, you can manually buy a 30-day credit pack: pay $9 for $10, $19 for $23 (Recommended), or $39 for $48. Pack credits work on all five models, never auto-renew, and remain valid for 30 days from purchase.
Our Claude and Gemini models cost 20%–40% less than official list rates: Claude Opus 5.5 is $2.80/$14 per 1M tokens (30% off), Sonnet 5.5 is $1.60/$8 (20% off), Gemini 4 Argon is $1.20/$6 (40% off), and Gemini 3.8 Flash is $0.49/$2.45 (~35% off). DeepSeek V4 Pro runs on the official DeepSeek API at a single flat rate 24/7 ($1.49/$4.49, close to official peak pricing).
Our server places cache breakpoints on your system/preset prefix, character card, and conversation history so upstream providers reuse those tokens at a 90%–95% discount on subsequent replies. In a typical 50-reply, 32k-context session on Claude Opus 5.5 (~1.6M input and 20k output tokens), caching lowers the total cost from roughly $4.76 to about $1.03.
New accounts receive $0.30 in free trial credit automatically after verifying an email address or signing in with a verified Google or Discord account. You can use the trial credit immediately on Gemini 3.8 Flash (gemini-3.8-flash) and Gemini 4 Argon (gemini-4-argon).