Parasail models
Serverless open-model inference with public per-token rates and cache discounts.
DeepSeek V4 Flash (Parasail)
$0.14 / $0.28 per 1M tokens
GLM-5.2 (Parasail)
$1.40 / $4.40 per 1M tokens
DeepSeek V4 Pro (Parasail)
$1.74 / $3.48 per 1M tokens
MiniMax M3 (Parasail)
$0.3 / $1.20 per 1M tokens
Mistral Small 3.2 24B (Parasail)
$0.09 / $0.3 per 1M tokens
gpt-oss-120b (Parasail)
$0.1 / $0.75 per 1M tokens
Gemma 4 31B (Parasail)
$0.15 / $0.4 per 1M tokens
Kimi K2.6 (Parasail)
$0.75 / $3.50 per 1M tokens
Parasail
Serverless open-model inference with public per-token rates and cache discounts.
There are 8 published models here. On a 70/30 mix, combined list rates run from $0.153 to $2.30 per 1 million tokens, with a median of $0.4325. The largest listed context window in this group is 1,000,000 tokens. The most recent price check in this group is 2026-08-14.
Official Parasail pricing page
Parasail list rates
USD per million tokens. The 70/30 mix is a planning default — change it on the model page if your traffic is output-heavy.
| Model | Input / 1M | Output / 1M | Mixed 70/30 | Context | Verified |
|---|---|---|---|---|---|
| Mistral Small 3.2 24B (Parasail) | $0.09 | $0.3 | $0.153 | 128,000 | |
| DeepSeek V4 Flash (Parasail) | $0.14 | $0.28 | $0.182 | 1,000,000 | |
| Gemma 4 31B (Parasail) | $0.15 | $0.4 | $0.225 | 128,000 | |
| gpt-oss-120b (Parasail) | $0.1 | $0.75 | $0.295 | 128,000 | |
| MiniMax M3 (Parasail) | $0.3 | $1.20 | $0.57 | 200,000 | |
| Kimi K2.6 (Parasail) | $0.75 | $3.50 | $1.575 | 256,000 | |
| DeepSeek V4 Pro (Parasail) | $1.74 | $3.48 | $2.262 | 1,000,000 | |
| GLM-5.2 (Parasail) | $1.40 | $4.40 | $2.30 | 200,000 |
Questions about Parasail rates
Do these Parasail figures include discounts and tax?
No. They are public list rates. Commits, credits, regions, tax, and commercial discounts are not in the number — check the source linked on each model.
When is the more expensive Parasail tier worth the extra?
GLM-5.2 (Parasail) sits near $2.30 per 1M tokens on a 70/30 mix, versus $0.153 on Mistral Small 3.2 24B (Parasail). Use the high tier when retries or long reasoning actually fail on the cheap one.