NewHow we catch a seller listing the wrong model, and refund you if one does.

Is there a free LLM API?

MartLLM has no free tier: accounts start at $0 and are prepaid, from $5, with no monthly fee and no minimum spend. On November 20, 2026, 124 chat models on MartLLM cost under $0.10 per million input tokens. A request that fails is not billed. Credit does not expire while the account is open, and unused credit can be refunded on request within 30 days of purchase.

Prices as of November 20, 2026, 11:00 UTC. This page is rebuilt every hour.

ModelInput / 1M tokensOutput / 1M tokensBelow list price
GPT-OSS 20B$0.000018$0.00008999.9%
GPT-OSS 120B$0.000069$0.00029799.9%
GPT OSS Safeguard 20B$0.000075$0.000399.9%
Qwen: Qwen3.8 Omni Flash$0.000146$0.00045699.9%
GLM 4.5 Air$0.000129$0.00084299.9%
Voxtral Mini 3B 2507$0.0004$0.000499%
Gemma 3 4B IT$0.000463$0.00057999%
Gemma 4 E2B$0.0004$0.000899%
MiniMax M2.1$0.000356$0.00143
Gemma 3 12B IT$0.000495$0.0014999.2%
GLM 4.7 Flash$0.000326$0.00299.5%
Gemma 4 26B A4B IT$0.0008$0.00199.2%
Ministral 3B$0.001$0.00199%
NVIDIA Nemotron Nano 9B v2$0.0006$0.002399%
Gemma 3 27B$0.000792$0.0019899%
Qwen3 Coder 30B A3B Instruct$0.000693$0.0027799.4%
Qwen3 32B$0.000792$0.0027799.3%
Qwen 3 235B A22B Instruct 2507$0.000567$0.0034799.5%
NVIDIA Nemotron 3 Nano 30B$0.000743$0.0029799%
Gemma 4 31B IT$0.000912$0.0027499.2%

MartLLM adds no fee to these prices · every model and price →

More questions

Is there a trial credit?

No. New accounts start at $0; the smallest top-up is $5, by card or PayPal.

What does a failed request cost?

Nothing. A request or a generation that fails costs $0; you are billed only for results you receive.