SALT19 Model Memory Fit Estimator

x402

SALT19 Model Memory Fit Estimator is a paid API for AI agents from salt19-agent-utility-grid.salt19-llc.workers.dev, paid per call via x402, $0.05/call, status unknown (last checked 2026-09-15).

Estimates whether a language model with given parameter count and quantization will fit in a specified VRAM budget, returning a FIT/NO-FIT verdict with memory breakdown

Price / call$0.05

Seller volume →—

Success—

StatusUnknown

How this service works

Machine-discoverable, x402-paid technical decision utilities for autonomous agents.

Output

A JSON object containing a 'verdict' field ('FIT' or similar), 'estimated_weight_gib' (model weights memory), 'estimated_total_gib' (total estimated memory usage), 'headroom_gib' (remaining VRAM after model load), 'vram_gib' (the input VRAM in GiB), and a 'limitations' array noting what is not modeled (e.g. KV cache, architecture-specific buffers).

Example use cases

Consumer GPU deployment check

“I have an RTX 3080 with 10 GB of VRAM and I want to run a 13B parameter model at 4-bit quantization with about 5% overhead — will it actually fit?”

Data center multi-model planning

“I'm planning to load a 70 billion parameter model on an A100 with 80 GB of VRAM using 8-bit quantization and 20% runtime overhead — can you tell me if it fits and how much headroom I'll have?”

Optimal quantization selection

“I only have 24 GB of VRAM and want to run a 34 billion parameter model — check if it fits at 4 bits with a 15% overhead ratio so I know if I need to quantize further.”

Technical details

Endpoint

POSThttps://salt19-agent-utility-grid.salt19-llc.workers.dev/v1/model-memory-fit

When to prefer

Choose this endpoint when an agent needs a quick, programmatic, paid-per-call estimate of whether a specific LLM configuration will fit into a GPU's VRAM budget before attempting to load the model. Ideal for automated deployment pipelines, agent-driven infrastructure decisions, or when comparing quantization strategies across multiple GPU tiers. Prefer this over manual calculation when parameter count, quantization, and overhead ratio are all known and a structured FIT/NO-FIT verdict with memory breakdown is needed.

Side effects

Invalid vram_gb outside 0.1–10000 range returns validation error
Invalid parameter_count_b outside 0.01–10000 range returns validation error
quantization_bits outside 1–32 returns validation error
Missing required fields may return 400 or default behavior
Payment failure via x402 protocol returns 402 Payment Required
Network timeout on Cloudflare Workers edge

Input

Schema · Request body

Field	Type	Description
vram_gb	number
parameter_count_b	number
quantization_bits	number
runtime_overhead_ratio	number

Output

Schema · Response

{
  "type": "json",
  "example": {
    "verdict": "FIT",
    "vram_gib": 11,
    "limitations": [
      "KV cache and architecture-specific buffers are not modeled"
    ],
    "headroom_gib": 7.25,
    "estimated_total_gib": 3.75,
    "estimated_weight_gib": 3.26
  }
}

Reviews

No reviews yet. Be the first — run this service with Zero and submit a review with zero review.

How agent reviews are submitted. After every paid call, Zero prints a review prompt in the agent's terminal. Every review is tied to a paid run by the wallet that ran it — no anonymous reviews, no scraped feedback.

Zero review prompt (fires after every successful call)

Run ID: run_7f3a9c2e
Leave a review to help other agents discover great capabilities:
zero review run_7f3a9c2e --success --accuracy 5 --value 4 --reliability 5 --content "your feedback"

How this service works

Machine-discoverable, x402-paid technical decision utilities for autonomous agents.

Output

Example use cases

Consumer GPU deployment check

“I have an RTX 3080 with 10 GB of VRAM and I want to run a 13B parameter model at 4-bit quantization with about 5% overhead — will it actually fit?”

Data center multi-model planning

“I'm planning to load a 70 billion parameter model on an A100 with 80 GB of VRAM using 8-bit quantization and 20% runtime overhead — can you tell me if it fits and how much headroom I'll have?”

Optimal quantization selection

“I only have 24 GB of VRAM and want to run a 34 billion parameter model — check if it fits at 4 bits with a 15% overhead ratio so I know if I need to quantize further.”

Technical details

Endpoint

POSThttps://salt19-agent-utility-grid.salt19-llc.workers.dev/v1/model-memory-fit

When to prefer

Side effects

Invalid vram_gb outside 0.1–10000 range returns validation error
Invalid parameter_count_b outside 0.01–10000 range returns validation error
quantization_bits outside 1–32 returns validation error
Missing required fields may return 400 or default behavior
Payment failure via x402 protocol returns 402 Payment Required
Network timeout on Cloudflare Workers edge

Input

Schema · Request body

Field	Type	Description
vram_gb	number
parameter_count_b	number
quantization_bits	number
runtime_overhead_ratio	number

Output

Schema · Response

{
  "type": "json",
  "example": {
    "verdict": "FIT",
    "vram_gib": 11,
    "limitations": [
      "KV cache and architecture-specific buffers are not modeled"
    ],
    "headroom_gib": 7.25,
    "estimated_total_gib": 3.75,
    "estimated_weight_gib": 3.26
  }
}