LiteLLMLiteLLM/ MODELS

ZeroGPU model pricing

As of October 31, 2026, 13 ZeroGPU models are available through LiteLLM. The most recently added are Lfm2.5 1.2B Instruct, Lfm2.5 1.2B Thinking, Llama 3.1 8B Instruct Fast. Prices below are ZeroGPU's list prices.

Prices updated Checked How to call ZeroGPU with LiteLLM

Cheapest chat models on ZeroGPU

Sorted by ZeroGPU's own prices, blended at 3 input tokens per output token. Active chat models only.

#ModelBlended per 1MInput / output per 1MContext
1Lfm2.5 1.2B Instruct$0.0275$0.02 / $0.0532K
2Lfm2.5 1.2B Thinking$0.0275$0.02 / $0.0532K
3Llama 3.2 3B Instruct Fast$0.1125$0.05 / $0.30128K
4GLM 5.3 Flash$0.1625$0.10 / $0.351M
5Llama 3.1 8B Instruct Fast$0.1825$0.15 / $0.28128K
6Qwen3 30B A3B$0.1875$0.10 / $0.4532K
7GPT-6 Luna$0.20$0.10 / $0.50272K
8GPT OSS 120B$0.2625$0.15 / $0.60128K
9GPT-5.6 Luna$0.45$0.20 / $1.20272K
10DeepSeek V4.1 Flash$0.525$0.30 / $1.201M

ZeroGPU models

ModelLabPriceContextProvidersAdded
Lfm2.5 1.2B Instruct–$0.02 / $0.05 per 1M tokens32K1Oct 2026
Lfm2.5 1.2B Thinking–$0.02 / $0.05 per 1M tokens32K1Oct 2026
Llama 3.1 8B Instruct FastMeta$0.15 / $0.28 per 1M tokens128K1Oct 2026
Llama 3.2 3B Instruct FastMeta$0.05 / $0.30 per 1M tokens128K1Oct 2026
GPT-6 LunaOpenAI$0.10 / $0.50 per 1M tokens272K12Sep 2026
DeepSeek V4.1 FlashDeepSeek$0.30 / $1.20 per 1M tokens1M22Sep 2026
GLM 5.3 FlashZ.ai$0.10 / $0.35 per 1M tokens1M21Aug 2026
GPT-5.6 LunaOpenAI$0.20 / $1.20 per 1M tokens272K12Jul 2026
Qwen3.6 35B A3BAlibaba$0.20 / $1.50 per 1M tokens256K10Jun 2026
GPT-5.4 NanoOpenAI$0.20 / $1.25 per 1M tokens272K8Mar 2026
GPT OSS 120BOpenAI$0.15 / $0.60 per 1M tokens128K27Aug 2025
Qwen3 30B A3BAlibaba$0.10 / $0.45 per 1M tokens32K10Jul 2025
GPT-4.1 MiniOpenAI$0.40 / $1.60 per 1M tokens1M5Apr 2025