Cheapest AI models that are still good

1. 1

   Claude Haiku 5.5 · Anthropic

   Ten cents a million and, per the people using it, smarter than the other ten-cent model; mind the 100K cliff.

   Score 74Index 43$0.10 in · $0.50 out per 1M1M context

   Caveat: Price jumps five-fold above 100K input tokens, which commenters call an absurdly low cutoff for agent runs.

2. 2

   GPT-6 Luna · OpenAI

   OpenAI's ten-cent model: the one every ChatGPT Free user now has, and the one Haiku users keep comparing themselves to.

   Score 66Index 38$0.10 in · $0.50 out per 1M1M context
3. 3

   DeepSeek Flash (4.1) · DeepSeek

   The reason a thousand people argued on Hacker News this week: frontier-adjacent coding for pocket change, with the caveats you would expect.

   Score 70$0.30 in · $1.20 out per 1M1M context

   Caveat: Open-weights status and licence not verified from a primary source at time of writing.

4. 4

   Muse Spark 1.3 · Meta

   The fastest thing in the table by a distance, and the one to beat on the stress benchmark nobody else talks about.

   Score 72Index 48$0.78 per 1M blended (est.)
5. 5

   Gemini 4 Argon · Google

   Wins the popularity contest: first on Arena, mid-pack on the index, and priced like a workhorse.

   Score 85Index 53$1.99 per 1M blended (est.)1M context

   Caveat: Google's public pricing page did not list Argon when read; the blended figure is Artificial Analysis' estimate.

6. 6

   Kimi K3 · Moonshot

   The highest-placed open-licence model on Arena, and the one most teams have not tried yet.

   Score 60Index 44$2.31 per 1M blended (est.)

   Caveat: Price is Artificial Analysis' blended figure, not a vendor list price; context not verified from a primary source.

7. 7

   Grok 4.7 · xAI

   Mid-table on the index at a frontier price, with a smaller context window than everyone above it.

   Score 58Index 46$3.74 per 1M blended (est.)500K context

   Caveat: 500K context is the smallest in the table.

8. 8

   Claude Sonnet 5.5 · Anthropic

   The value pick of the table: second on the index at a fifth of the Opus price, and the fastest of the top five.

   Score 84Index 56$2 in · $10 out per 1M1M context
9. 9

   GPT-6.1 Sol · OpenAI

   Near-Astra intelligence at a fifth of the price, which is OpenAI's own line and, for once, the index agrees.

   Score 82Index 52$2 in · $10 out per 1M1M context
10. 10

    Claude Opus 5.5 · Anthropic

    The model people actually ship agents on: top of the independent index, a price that does not need a budget meeting, and a community that complains mainly about how much it has to be used.

    Score 91Index 58$4 in · $20 out per 1M1M context

    Caveat: Arena puts Gemini 4 Argon ahead on chat preference; the gap is inside two scores' error bars.

11. 11

    GPT-6 Astra · OpenAI

    The reasoning heavyweight: leads the hardest exams and the cipher-cracking party tricks, and charges like it.

    Score 86Index 53$10 in · $50 out per 1M1M context

    Caveat: Long-context pricing doubles above 272K input tokens.

12. 12

    Claude Fable 5.1 · Anthropic

    The specialist: Anthropic's own docs say to reach for it when Opus 5.5 at high effort still falls short, and the price says the same.

    Score 80Index 53$10 in · $50 out per 1M1M context

    Caveat: Ten dollars per million input is the highest in this table alongside Astra.

Every figure links to its source on the model profile pages. Vendor figures are labelled as vendor figures and never counted as independent.