Cheapest open-weight LLMs
Mistral: Mistral Nemo is the cheapest option at $0.029 per million input tokens. Models whose weights you can download and run yourself. The API price here is what you pay to avoid operating the hardware.
158 models qualify
Ranked by blended cost - input weighted 75%, output 25%.
| # | Model | Blended / 1M | Input / 1M | Output / 1M | Context | Capabilities |
|---|---|---|---|---|---|---|
| 1 | Flint (free) Recursive | Free | Free | Free | 262K | Free tierReasoningToolsOpen weights |
| 2 | Xing4.0 29B A4B (free) XingChen | Free | Free | Free | 262K | Free tierReasoningToolsOpen weights |
| 3 | North Mini Code (free) Cohere | Free | Free | Free | 256K | Free tierReasoningToolsOpen weights |
| 4 | Free | Free | Free | 256K | Free tierReasoningToolsVisionOpen weights | |
| 5 | Mistral Nemo Mistral AI | $0.029 | $0.029 | $0.030 | 131K | ToolsOpen weights |
| 6 | Ling 3.0 Flash inclusionAI | $0.032 | $0.021 | $0.063 | 262K | ReasoningToolsOpen weights |
| 7 | gpt-oss-20b OpenAI | $0.036 | $0.018 | $0.090 | 131K | ReasoningToolsOpen weights |
| 8 | Llama 3 8B Lunaris Sao10K | $0.042 | $0.040 | $0.050 | 8K | Open weights |
| 9 | Nex-N2.5-Mini Nex AGI | $0.044 | $0.025 | $0.100 | 262K | ReasoningVisionOpen weights |
| 10 | GLM 5.3 Flash Z.ai | $0.053 | $0.046 | $0.074 | 1.0M | ReasoningToolsVisionOpen weights |
| 11 | Mistral Small 3 Mistral AI | $0.057 | $0.050 | $0.080 | 33K | Open weights |
| 12 | Llama 3.1 8B Instruct Meta Llama | $0.057 | $0.050 | $0.080 | 131K | ToolsOpen weights |
| 13 | Schematron V2 Turbo Inference.net | $0.060 | $0.030 | $0.150 | 128K | Open weights |
| 14 | Gemma 3 4B | $0.063 | $0.050 | $0.100 | 131K | VisionOpen weights |
| 15 | Llama 3.2 1B Instruct Meta Llama | $0.071 | $0.027 | $0.201 | 60K | Open weights |
| 16 | Laguna XS 2.1 Poolside | $0.075 | $0.060 | $0.120 | 262K | Free tierReasoningToolsOpen weights |
| 17 | Gemma 3 12B | $0.075 | $0.050 | $0.150 | 131K | ToolsVisionOpen weights |
| 18 | Hy-MT2-1.8B Tencent | $0.077 | $0.044 | $0.177 | 8K | Open weights |
| 19 | Granite 4.2 8B IBM Granite | $0.082 | $0.060 | $0.150 | 131K | ReasoningToolsOpen weights |
| 20 | MythoMax 13B MythoMax 13B | $0.087 | $0.080 | $0.110 | 8K | Open weights |
| 21 | Phi 4 Microsoft | $0.088 | $0.070 | $0.140 | 16K | Open weights |
| 22 | Ling 3.0 Flash VL inclusionAI | $0.090 | $0.060 | $0.180 | 131K | ReasoningToolsVisionOpen weights |
| 23 | Ternary Bonsai 2 9B PrismML | $0.092 | $0.080 | $0.130 | 262K | ReasoningToolsVisionOpen weights |
| 24 | Schematron V2 Small Inference.net | $0.095 | $0.050 | $0.230 | 128K | Open weights |
| 25 | Reka Edge Reka AI | $0.100 | $0.100 | $0.100 | 16K | ToolsVisionOpen weights |
Frequently asked
What is the cheapest LLM for self-hostable models?
Mistral: Mistral Nemo at $0.029 per million input tokens and $0.030 per million output tokens, with a 131K token context window.
How many models qualify for self-hostable models?
158 of the 359 models tracked here meet the criteria for this list.
How much do prices vary within this category?
By roughly 150×, from Mistral Nemo at the bottom to Command A at the top.