AGPL-3.0-or-later · forever.

the models · 65

Every model, one room.

The union of the wire catalog: each model names the providers that serve it, the window it reads, the exact-match price rows and the measured energy rows. What the catalog does not carry, the room says plainly · a missing number is a fact, never a zero.

Vendored from the released binary · supernovae-st/nika@f3a31a6ee007 (v0.118.7) · digest-verified, re-derived at build, gated in CI

The bench

1 published receipt: a room shows a measured answer only when a receipt names its model, with the evidence pack served beside it. Every other seat: run it yourself · the model-bench workflow in the public registry measures YOUR machine, and the first receipt shows what an honest one looks like.

The context axis

The axis draws the canonical catalog · the spec-named seats whose models declare a window. The register below is the wider wire union; each count names its facet.

Loading the context axis…

The price axis

Every priced model, seated at its cheapest recorded output seat. The taller accent ticks are open weights: the seats you could also host yourself.

13 priced · $0.08 → $15 out/Mtok · log scale · 6 open weights

The register

  1. cloudflare

  2. cloudflare

  3. sambanova

  4. sambanova

  5. minimax

  6. minimax

  7. native · vllm

  8. huggingface

  9. huggingface

  10. fireworks

  11. fireworks

  12. bedrock

  13. bedrock

  14. openrouter

  15. openrouter

  16. anthropic

  17. anthropic

  18. cohere

  19. cohere

  20. databricks

  21. deepseek-chat64k$0.28open weights

    deepseek

  22. llamacpp · localai

  23. mock

  24. gemini · vertex

  25. gemini · vertex

  26. zhipu

  27. zhipu

  28. gpt-4o128k

    azure

  29. gpt-4o-mini128k$0.6

    azure · openai

  30. openai

  31. gpt-5.2272k$14

    openai

  32. grok-3131k$15

    xai

  33. xai

  34. ai21

  35. ai21

  36. moonshot

  37. moonshot

  38. cerebras

  39. llama-3.1-8b-instant128k$0.08open weights

    groq

  40. cerebras

  41. llama-3.3-70b-versatile128k$0.79open weights

    groq

  42. ollama

  43. together

  44. hyperbolic

  45. together

  46. deepinfra

  47. deepinfra

  48. nvidia

  49. replicate

  50. replicate

  51. mistral-large-latest128k$1.5open weights

    mistral

  52. mistral-small-latest32k$0.6open weights

    mistral

  53. mock

  54. nvidia

  55. nvidia

  56. writer

  57. qwen

  58. qwen

  59. qwen

  60. lmstudio

  61. ollama

  62. sonar128k

    perplexity

  63. perplexity

  64. voyage

  65. voyage