Price list · USD per 1M tokens
Cheapest LLM API — from $0.017 per 1M tokens
Every open-weight model we serve on dedicated GPUs we operate, cheapest first. One OpenAI-compatible API, zero data retention, batch at half price.
| Model | Input / output | Batch (24h) | Context | vs OpenRouter endpoints |
|---|---|---|---|---|
gpt-oss-20bgpt-oss-20b | $0.017 / $0.085 | $0.0085 / $0.0425 | 65,536 | Lower on input than all 11 ($0.018 in) |
Gemma 4 26Bgemma-4-26b-a4b-it | $0.04 / $0.21 | $0.02 / $0.105 | 32,768 | Lowest there: $0.013 / $0.038 |
Qwen3 Coder 30Bqwen3-coder-30b | $0.0675 / $0.27 | $0.03375 / $0.135 | 98,304 | Lower than all 2 ($0.07 / $0.28) |
Qwen3 30B A3B Instruct 2507qwen/qwen3-30b-a3b-instruct-2507 | $0.0813 / $0.2438 | $0.04063 / $0.1219 | 98,304 | Lower than all 3 ($0.09 / $0.30) |
Qwen3 VL 30Bqwen/qwen3-vl-30b-a3b-instruct | $0.14 / $0.55 | $0.07 / $0.275 | 65,536 | Lower than all 3 ($0.15 / $0.60) |
Muse Glimmer 30Bmeta/muse-glimmer-30b | $0.28 / $1.10 | $0.14 / $0.55 | 131,072 | Lower than all 1 ($0.30 / $1.20) |
USD per 1M tokens. Synap prices are read live from /v1/models, which is what the gateway bills, and refresh on this page when it loads. Batch prices apply to jobs sent through /v1/batches with a 24-hour window. OpenRouter list prices checked 2026-11-11 15:36 UTC, lowest input and lowest output across every endpoint listing the model: gpt-oss-20b · Gemma 4 26B · Qwen3 Coder 30B · Qwen3 30B A3B Instruct 2507 · Qwen3 VL 30B · Muse Glimmer 30B.
How to pick
- Lowest cost per token for classification, extraction and routing: gpt-oss-20b
- Code, agents in an editor and tool calls: Qwen3 Coder 30B
- Screenshots, photos and scanned documents: Qwen3 VL 30B
- Not sure? Send
synap-v1as the model and the router picks the cheapest of these that can do the job, billed at the price of the model that ran.
Switch in one line
from openai import OpenAI
client = OpenAI(base_url="https://synap402.com/v1", api_key="sk-pool-...")
r = client.chat.completions.create(model="gpt-oss-20b", messages=[{"role": "user", "content": "Hello"}])Any OpenAI SDK works. Tool setup for Claude Code, Codex, Cline and more: guides.
How you pay
- Card from $10; credits you buy never expire
- USDC on Base through x402, no account: the paying wallet is the account (x402 guide)
- Try any model in chat with no sign-up
Larger models (GLM, DeepSeek, Qwen3.5 122B) are also in the catalogue; see all models and pricing.