Home › Trends › Cheapest LLM APIs per million tokens
Cheapest LLM APIs per million tokens
The cheapest paid LLM APIs on OpenRouter, ranked by price per million tokens from today's snapshot (27 Oct 2026) of 464 models. 20 free-tier variants are excluded. Blended price assumes 3 input tokens per output token.
Cheapest 30 by blended price
| # | Model | Input $/1M | Output $/1M | Blended $/1M (3:1) | Context | 7-day change |
|---|---|---|---|---|---|---|
| 1 | mistralai/mistral-nemo | $0.029 | $0.03 | $0.0293 | 131,072 | 0% |
| 2 | scaledown/summarize | $0.04 | $0 | $0.03 | 1,000,000 | 0% |
| 3 | scaledown/extract | $0.04 | $0 | $0.03 | 1,000,000 | 0% |
| 4 | inclusionai/ling-3.0-flash | $0.021 | $0.063 | $0.0315 | 262,144 | 0% |
| 5 | openai/gpt-oss-20b | $0.018 | $0.09 | $0.036 | 131,072 | 0% |
| 6 | google/gemma-3-4b-it | $0.04 | $0.05 | $0.0425 | 131,072 | 0% |
| 7 | sao10k/l3-lunaris-8b | $0.04 | $0.05 | $0.0425 | 8,192 | 0% |
| 8 | nex-agi/nex-n2.5-mini | $0.025 | $0.1 | $0.0438 | 262,144 | 0% |
| 9 | openai/gpt-oss-20b:batch | $0.024 | $0.112 | $0.046 | 131,072 | 0% |
| 10 | qwen/qwen3.7-flash | $0.03 | $0.13 | $0.055 | 1,000,000 | 0% |
| 11 | ~z-ai/glm-flash-latest | $0.0513 | $0.0701 | $0.056 | 1,048,576 | -26% |
| 12 | openai/gpt-oss-120b:batch | $0.0296 | $0.136 | $0.0562 | 131,072 | 0% |
| 13 | mistralai/mistral-small-24b-instruct-2501 | $0.05 | $0.08 | $0.0575 | 32,768 | 0% |
| 14 | meta-llama/llama-3.1-8b-instruct | $0.05 | $0.08 | $0.0575 | 131,072 | 0% |
| 15 | inference-net/schematron-v2-turbo | $0.03 | $0.15 | $0.06 | 128,000 | 0% |
| 16 | amazon/nova-micro-v1 | $0.035 | $0.14 | $0.0613 | 128,000 | 0% |
| 17 | cohere/command-r7b-12-2024 | $0.0375 | $0.15 | $0.0656 | 128,000 | 0% |
| 18 | inception/mercury-2.5 | $0.04 | $0.15 | $0.0675 | 260,000 | 0% |
| 19 | openai/gpt-5-nano:batch | $0.025 | $0.2 | $0.0688 | 400,000 | 0% |
| 20 | meta-llama/llama-3.2-1b-instruct | $0.027 | $0.201 | $0.0705 | 60,000 | 0% |
| 21 | ~deepseek/deepseek-v4-flash-latest | $0.0475 | $0.1568 | $0.0748 | 1,048,576 | -36% |
| 22 | poolside/laguna-xs-2.1 | $0.06 | $0.12 | $0.075 | 262,144 | 0% |
| 23 | deepseek/deepseek-v4.1-flash:batch | $0.05 | $0.15 | $0.075 | 1,048,576 | 0% |
| 24 | google/gemma-3-12b-it | $0.05 | $0.15 | $0.075 | 131,072 | 0% |
| 25 | tencent/hy-mt2-1.8b | $0.044 | $0.177 | $0.0772 | 8,192 | 0% |
| 26 | z-ai/glm-5.3-flash:batch | $0.05 | $0.18 | $0.0825 | 1,048,576 | 0% |
| 27 | gryphe/mythomax-l2-13b | $0.08 | $0.11 | $0.0875 | 8,192 | 0% |
| 28 | upstage/solar-mini4 | $0.05 | $0.2 | $0.0875 | 524,288 | 0% |
| 29 | openai/gpt-4.1-nano:batch | $0.05 | $0.2 | $0.0875 | 1,047,576 | 0% |
| 30 | microsoft/phi-4 | $0.07 | $0.14 | $0.0875 | 16,384 | 0% |
Cheapest 15 with 200k+ context
| # | Model | Input $/1M | Output $/1M | Blended $/1M (3:1) | Context | 7-day change |
|---|---|---|---|---|---|---|
| 1 | scaledown/summarize | $0.04 | $0 | $0.03 | 1,000,000 | 0% |
| 2 | scaledown/extract | $0.04 | $0 | $0.03 | 1,000,000 | 0% |
| 3 | inclusionai/ling-3.0-flash | $0.021 | $0.063 | $0.0315 | 262,144 | 0% |
| 4 | nex-agi/nex-n2.5-mini | $0.025 | $0.1 | $0.0438 | 262,144 | 0% |
| 5 | qwen/qwen3.7-flash | $0.03 | $0.13 | $0.055 | 1,000,000 | 0% |
| 6 | ~z-ai/glm-flash-latest | $0.0513 | $0.0701 | $0.056 | 1,048,576 | -26% |
| 7 | inception/mercury-2.5 | $0.04 | $0.15 | $0.0675 | 260,000 | 0% |
| 8 | openai/gpt-5-nano:batch | $0.025 | $0.2 | $0.0688 | 400,000 | 0% |
| 9 | ~deepseek/deepseek-v4-flash-latest | $0.0475 | $0.1568 | $0.0748 | 1,048,576 | -36% |
| 10 | poolside/laguna-xs-2.1 | $0.06 | $0.12 | $0.075 | 262,144 | 0% |
| 11 | deepseek/deepseek-v4.1-flash:batch | $0.05 | $0.15 | $0.075 | 1,048,576 | 0% |
| 12 | z-ai/glm-5.3-flash:batch | $0.05 | $0.18 | $0.0825 | 1,048,576 | 0% |
| 13 | upstage/solar-mini4 | $0.05 | $0.2 | $0.0875 | 524,288 | 0% |
| 14 | openai/gpt-4.1-nano:batch | $0.05 | $0.2 | $0.0875 | 1,047,576 | 0% |
| 15 | anthropic/claude-haiku-5.5:batch | $0.05 | $0.25 | $0.1 | 1,000,000 | 0% |
Cheapest 15 by output price
| # | Model | Input $/1M | Output $/1M | Blended $/1M (3:1) | Context | 7-day change |
|---|---|---|---|---|---|---|
| 1 | scaledown/summarize | $0.04 | $0 | $0.03 | 1,000,000 | 0% |
| 2 | scaledown/extract | $0.04 | $0 | $0.03 | 1,000,000 | 0% |
| 3 | mistralai/mistral-nemo | $0.029 | $0.03 | $0.0293 | 131,072 | 0% |
| 4 | google/gemma-3-4b-it | $0.04 | $0.05 | $0.0425 | 131,072 | 0% |
| 5 | sao10k/l3-lunaris-8b | $0.04 | $0.05 | $0.0425 | 8,192 | 0% |
| 6 | inclusionai/ling-3.0-flash | $0.021 | $0.063 | $0.0315 | 262,144 | 0% |
| 7 | ~z-ai/glm-flash-latest | $0.0513 | $0.0701 | $0.056 | 1,048,576 | -26% |
| 8 | mistralai/mistral-small-24b-instruct-2501 | $0.05 | $0.08 | $0.0575 | 32,768 | 0% |
| 9 | meta-llama/llama-3.1-8b-instruct | $0.05 | $0.08 | $0.0575 | 131,072 | 0% |
| 10 | openai/gpt-oss-20b | $0.018 | $0.09 | $0.036 | 131,072 | 0% |
| 11 | nex-agi/nex-n2.5-mini | $0.025 | $0.1 | $0.0438 | 262,144 | 0% |
| 12 | rekaai/reka-edge | $0.1 | $0.1 | $0.1 | 16,384 | 0% |
| 13 | mistralai/ministral-3b-2512 | $0.1 | $0.1 | $0.1 | 131,072 | 0% |
| 14 | gryphe/mythomax-l2-13b | $0.08 | $0.11 | $0.0875 | 8,192 | 0% |
| 15 | openai/gpt-oss-20b:batch | $0.024 | $0.112 | $0.046 | 131,072 | 0% |
7-day change compares the blended price with our 20 Oct 2026 snapshot ("new" = not listed then). Prices are OpenRouter list prices; provider-direct prices can differ.
Get the data behind this page
Every OpenRouter price change, new model and removal since 14 Oct 2026 is in our free CSV download (updated 27 Oct 2026). For daily JSON, full history and change events by API or MCP (watch any model, repo or MCP server and get an email when it moves), see plans from $19/month, or try the free sample.
Agent Data Weekly
The week's notable AI agents, MCP servers, launches and price changes, from our daily archive. Free, double opt-in, unsubscribe any time. Past issues
Related: DeepSeek vs Kimi API pricing · OpenRouter price changes this week · LLM API pricing by provider · New LLMs on OpenRouter this week · Trending Hugging Face models this week · Fastest-growing AI GitHub repos this week · New MCP servers this week · Today's AI ecosystem movers