Claude Haiku 5.5 and GPT-6 Luna tie as the cheapest LLM API in October 2026: both list at $0.10 / $0.50 per million input and output tokens. Next come MiMo V2.6 Flash ($0.14 / $0.28), Qwen 3.8 Flash ($0.15 / $0.47) and GLM 5.3 Flash ($0.15 / $0.50). The ranking below uses each maker's own list price, never a promotion. Per-token pricing wins light use; for agents that run all day, the math flips toward a flat monthly plan, with the working shown further down.
US dollars per million tokens, read on each maker's own pricing page on Oct 14, 2026 (the maker name links to it). Standard rates only: where a maker runs a promotion, the table shows the price it returns to. Ranked by blended price at 10 input tokens per output token, the mix of the coding-agent example below; caching discounts not included. Cheapest per token is not cheapest per finished task: see cost per coding task and all 32 models in the LLM API pricing comparison.
If your agent runs minutes a day, don't pay a subscription for it. These are the current per-token prices (live from OpenRouter data), plus what our worked example (an active agent at 150 requests/day, ~54M input + 5.4M output tokens/month) would cost on each:
Prices from the OpenRouter catalog snapshot this site syncs regularly; caching discounts not included. Redo the math for your own workload with the live cost calculator.
The table above already shows the turn: at active-agent volume, even mid-priced per-token models cross $39/month, and frontier models cross hundreds. Run the same workload on frontier quality per-token (Sonnet-class at ~$3/$15) and the example costs roughly $243/month, for one agent. Flat-rate flips that: a fixed monthly price (from $19/month, with higher tiers buying execution speed), and the heavier your agent runs, the better the deal gets. The trade-off is honest too: smart routing picks the model per request unless you choose your own models, and requests stop when the monthly compute budget is used.
Claude Haiku 5.5 and GPT-6 Luna tie as the cheapest LLM API in October 2026: both list at $0.10 / $0.50 per million input and output tokens. Next come MiMo V2.6 Flash ($0.14 / $0.28), Qwen 3.8 Flash ($0.15 / $0.47) and GLM 5.3 Flash ($0.15 / $0.50). Those are the model makers' own list prices, checked Oct 14, 2026. DeepSeek V4.1 Flash lists at $0.30 / $1.20 at peak hours and half that off-peak. For sustained agent workloads, a flat monthly plan (from $19) caps the bill once your agent runs for hours a day.
Compute your monthly tokens. A light agent (a few dozen requests a day) costs single-digit dollars on cheap per-token models, and flat rate can't beat that. An active agent (150+ requests/day with real context) burns tens of millions of tokens monthly; on frontier per-token prices that's hundreds of dollars, where a flat plan at $39 wins by an order of magnitude.
Retry storms (failed and rate-limited requests still consume attempts and engineering time), context re-sending (agents resend history every step, multiplying input tokens), and bill variance (one runaway loop can cost more than a year of a flat plan). Cheap per-token pricing is real, but agents are built to consume it aggressively.
Free tiers are for evaluation, not operation: daily request caps and aggressive rate limits end agent sessions mid-task. They're great for trying a model; they're not an operating plan for anything always-on.