Cheapest LLM API
The cheapest LLM API on OceanToken is Llama 3.1 8B Instruct, at $0.05 / $0.08 per 1M tokens. Below, 20 models ranked by the price every account pays.
Ranked by price 20
| # | Model | Family | Context | Price | Discount |
|---|---|---|---|---|---|
| 1 | Llama 3.1 8B Instructmeta-llama/llama-3.1-8b-instruct | Llama | 131K | $0.05 / $0.08 per 1M tokens | — |
| 2 | Qwen3.7 Flashqwen/qwen3.7-flash | Qwen | 984K | $0.03 / $0.13 per 1M tokens | — |
| 3 | Nova Micro 1.0amazon/nova-micro-v1 | Nova | 128K | $0.035 / $0.14 per 1M tokens | — |
| 4 | Gemma 3 12Bgoogle/gemma-3-12b-it | Gemma | 131K | $0.05 / $0.15 per 1M tokens | — |
| 5 | Ministral 3 3B 2512mistralai/ministral-3b-2512 | Mistral | 131K | $0.10 / $0.10 per 1M tokens | — |
| 6 | Nemotron 3.5 Lightningnvidia/nemotron-3.5-lightning | Nemotron | 262K | $0.06 / $0.16 per 1M tokens | — |
| 7 | gpt-oss-20bopenai/gpt-oss-20b | GPT | 131K | $0.04 / $0.18 per 1M tokens | — |
| 8 | Qwen3.5-9Bqwen/qwen3.5-9b | Qwen | 262K | $0.10 / $0.15 per 1M tokens | — |
| 9 | Nova Lite 1.0amazon/nova-lite-v1 | Nova | 300K | $0.06 / $0.24 per 1M tokens | — |
| 10 | Ministral 3 8B 2512mistralai/ministral-8b-2512 | Mistral | 262K | $0.15 / $0.15 per 1M tokens | — |
| 11 | Mistral Nemomistralai/mistral-nemo | Mistral | 131K | $0.15 / $0.15 per 1M tokens | — |
| 12 | Nemotron 3 Nano 30B A3Bnvidia/nemotron-3-nano-30b-a3b | Nemotron | 262K | $0.06 / $0.24 per 1M tokens | — |
| 13 | Qwen2.5 7B Instructqwen/qwen-2.5-7b-instruct | Qwen | 33K | $0.10 / $0.20 per 1M tokens | — |
| 14 | Qwen3.5-Flashqwen/qwen3.5-flash-02-23 | Qwen | 984K | $0.065 / $0.26 per 1M tokens | — |
| 15 | Qwen3 Coder 30B A3B Instructqwen/qwen3-coder-30b-a3b-instruct | Qwen | 262K | $0.07 / $0.28 per 1M tokens | — |
| 16 | Qwen3 14Bqwen/qwen3-14b | Qwen | 41K | $0.12 / $0.24 per 1M tokens | — |
| 17 | Seed 1.6 Flashbytedance-seed/seed-1.6-flash | Seed | 262K | $0.075 / $0.30 per 1M tokens | — |
| 18 | gpt-oss-safeguard-20bopenai/gpt-oss-safeguard-20b | GPT | 131K | $0.075 / $0.30 per 1M tokens | — |
| 19 | Qwen3 30B A3B Instruct 2507qwen/qwen3-30b-a3b-instruct-2507 | Qwen | 262K | $0.09 / $0.30 per 1M tokens | — |
| 20 | Llama 4 Scoutmeta-llama/llama-4-scout | Llama | 1.31M | $0.10 / $0.30 per 1M tokens | — |
About these prices
Frequently asked questions
What is the cheapest LLM API?
On OceanToken it is Llama 3.1 8B Instruct, at $0.05 / $0.08 per 1M tokens. Next is Qwen3.7 Flash, at $0.03 / $0.13 per 1M tokens. Both answer on the same API with the same key.
How are the models ranked?
Ranked by blended price (half input, half output tokens, per 1M), as every account pays it; only models of named families are ranked.
Is the cheapest model good enough?
The price and the listed capabilities are facts this page can show; how well a model does your task is not. Run the same prompts on two models: with one key, switching is a one-line change.
Keep exploring
Llama API · Nova API · Gemma API · Mistral API · Nemotron API · GPT API · Seed API
Cheapest long-context LLM API · Cheapest vision LLM API · Cheapest reasoning model API · Cheapest video generation API · Cheapest image generation API · Cheapest embedding API