SN64Verified snapshot
Chutes
Per 1M tokens
$0.104 in · $0.416 out
Qwen3-32B TEE · high confidence
2026-08-30 verified snapshot
Official source ↗Bittensor now · FLOP next
Route inference with the evidence attached.
A neutral comparison surface for decentralized inference—pricing, availability, and source confidence in one auditable view.
Last source checkAUG 30, 2026Verified snapshot
5adapters
3priced surfaces
0credentials stored
Q4FLOP testnet target
Provider board
Current market signals
Units are preserved as published; unlike units are never ranked as if equivalent.
SN19Qualified headline
Nineteen
Per 1M tokens
$0.00 / 1M tokens*
Provider landing-page headline · medium confidence
Asterisked “currently” headline
Official source ↗SN110Verified snapshot
Green Compute
GPU rental
From $0.40 / GPU hr
RTX 4090 · high confidence
Indicative starting rate
Official source ↗SN46Integration watch
Instant
Public customer price
Not published
OpenAI-compatible inference · high confidence
Public price not published
Official source ↗Testnet target Q4 2026Adapter ready—not connected
FLOP
Testnet price
Not available yet
Official testnet inference · high confidence
Interface not released
Official source ↗Route estimate
Price a representative workload
The estimator uses the latest Chutes split-token quote. Active inference stays disabled until credentials are supplied through protected hosting configuration and an explicit spend cap is approved.
Current verifiable routeChutes · Qwen3-32B TEEOnly reviewed adapter with split token pricing for this model.
Estimated Chutes cost$0.0133Published rate · estimate only · no request sent
Benchmark contract
Useful activity, bounded by design.
When credentials are authorized, each run records cost, latency, throughput, model identity, source time, and a redacted evidence digest.
TTFTTime to first token
TOK/SOutput throughput
USDObserved total cost
SHA-256Redacted evidence digest
Wallet connection offPaid inference offFLOP adapter fail-closed