Inference · /inferences/seasons/2026-W39

Season 2026-W39, open.

21 Sept 2026 to 28 Sept 2026 — 7 days left. What the grams proved of the catalog this window, per artifact and hardware class. Lineage is from the catalog; every number is a signed measurement or a count of them, and a gram count is a verified floor.

0grams contributed
0artifacts proved
0hardware classes
0publishers
cohorts this season
  • 0 · Lead member, rung 00
  • 1 · Lead member's ladder0
  • 2 · Other vendor members0
  • 3 · Derivatives0

The order a season proves things in: the number every other number is compared to comes first.

hardware classes that signed

    A class is os · arch · card class · RAM band. A number proved on one says nothing about another.

    cells the network has signed as unprovable
    • bharatgenai/sooktam2Windows · x86-64 · Intel GPU · 16 GiB · unavailable · 1 gram

      the ref, or the quant asked of it, is not published upstream

    • coqui/XTTS-v2Windows · x86-64 · AMD · 8 GiB · unavailable · 1 gram

      the ref, or the quant asked of it, is not published upstream

    • coqui/XTTS-v2Windows · x86-64 · NVIDIA · 8 GiB · unavailable · 1 gram

      the ref, or the quant asked of it, is not published upstream

    • coqui/XTTS-v2macOS · ARM64 · Apple GPU · 16 GiB · unavailable · 1 gram

      the ref, or the quant asked of it, is not published upstream

    • ResembleAI/chatterboxmacOS · ARM64 · Apple GPU · 16 GiB · unavailable · 1 gram

      the ref, or the quant asked of it, is not published upstream

    • SWivid/F5-TTSmacOS · ARM64 · Apple GPU · 16 GiB · unavailable · 1 gram

      the ref, or the quant asked of it, is not published upstream

    refusals are signed facts about (artifact, class): gated, unloadable, or absent upstream. They are coverage of a kind -- the fleet learned it once -- and they sink the cell to the back of every gram of that class. A refusal lifts when its season closes and the engine that refused has moved on.

    what proving costs, per hardware class
    • Windows · x86-64 · Intel GPU · 16 GiB5 MiB/s median fetch
      • generation: 6 min 3 s median probe · 1 sample
    • Windows · x86-64 · NVIDIA · 16 GiB14.8 MiB/s median fetch
      • ollama: 17 min 42 s median probe · 1 sample
      • ollama-host: 27 min 35 s median probe · 1 sample

      carried forward — no gram of this class reported in the latest aggregate

    • macOS · x86-64 · Apple GPU · 8 GiB305.8 MiB/s median fetch

        carried forward — no gram of this class reported in the latest aggregate

      • Windows · x86-64 · AMD · 8 GiB894.6 MiB/s median fetch
        • ollama-host: 30 min 52 s median probe · 3 samples
        • generation: 1 s median probe · 1 sample
      • macOS · ARM64 · Apple GPU · 16 GiB11.3 MiB/s median fetch
        • ollama-host: 4 min 8 s median probe · 1 sample
      • Windows · x86-64 · NVIDIA · 8 GiB3.7 MiB/s median fetch
        • Windows · x86-64 · no GPU · 8 GiB7.3 MiB/s median fetch
          • generation: 1 s median probe · 1 sample
          • ollama-host: 30 min 52 s median probe · 3 samples
        • Windows · x86-64 · no GPU · 16 GiB18.4 MiB/s median fetch

            carried forward — no gram of this class reported in the latest aggregate

          cost is operational: medians of node-reported pull/probe clocks per hardware class and engine, with sample counts. It is not a measurement of any model and is never ranked.

          what the network actually used
          • gemma4:e2b213.2k tokens served
          • hf.co/unsloth/Qwen3-4B-GGUF:IQ3_XXS124.5k tokens served
          • claude-haiku-4-597.2k tokens served
          • hf.co/unsloth/Phi-4-mini-reasoning-GGUF:IQ1_S23.3k tokens served
          • hf.co/unsloth/gemma-4-E2B-it-GGUF:Q4_K_M20.6k tokens served
          • llama3.2:1b15.1k tokens served
          • hf.co/unsloth/Qwen3-4B-GGUF:IQ1_S8.0k tokens served
          • gemma4:e2b-it-qat7.8k tokens served
          • hf.co/Qwen/Qwen2.5-0.5B-Instruct-GGUF:q4_k_m7.1k tokens served
          • hf.co/Qwen/Qwen2.5-0.5B-Instruct-GGUF:Q4_K_M7.1k tokens served
          • hf.co/Qwen/Qwen3-4B-GGUF:Q5_07.0k tokens served
          • hf.co/unsloth/Phi-4-mini-reasoning-GGUF:IQ3_XXS5.3k tokens served

          demand is tokens produced per artifact per class, from signed throughput observations; it orders depth, never existence.

          Every number is a signed measurement or a count of them. grams is how many distinct nodes contributed: a verified floor, not a complete account. Lineage (publisher, attribution, base, cohort, rung) is from the catalog and is not a measurement.

          grams and classes are what THIS window heard from; coverage, refusals, cost and demand carry across windows, because a gram going to sleep does not unmake what it measured.

          The season file → All seasons →