INFERENCE DEMAND DATA

Inference Token Prices

Real-time inference economics across hosted-model providers - token prices, throughput, and latency as a demand-side read on AI compute.

2,397 QUOTES · 381 MODELSParquet · JSON · CSVContinuous polling, daily aggregates

2,397

Token-price quotes

381

Models priced

1,366

Utilization snapshots

Hourly

Collection cadence

Download Package Contents Build process Verification Audit trail Use cases Delivery Further reading Request a sample

01Download

Inspect a real sample

Representative records in the delivery format, ready to inspect before licensing the full dataset.

Current collected record (output price)

Representative. The study measures output price per million tokens on cheapest-available models; tiers are not pooled.

inference_prices.jsonlrepresentative

{
  "observed_at": "2026-04-02T00:00:00Z",
  "provider": "provider-a",
  "model_family": "llama",
  "model_listing": "llama-3.1-8b-instruct",
  "output_price_per_mtok_usd": 0.02,
  "input_price_per_mtok_usd": 0.01,
  "model_tier": "small_open"
}

Marketplace utilization snapshot

Real row shape from the reference sample. 1,366 snapshots, every one carrying a computed ratio.

provider-utilization.csvrepresentative

provider,sku_canonical,instance_total,instance_available,instance_rented,utilization_ratio,snapshot_ts
marketplace-a,RTX_3090_24GB,2368,1120,1248,0.527,2025-12-03T09:36:00Z

02Schema

Record shape

Every field, its type, whether it can be null, and a representative value.

observed_at

required

When the price was collected from the provider pricing surface.

Type: timestamp
Unit: UTC

Example 2026-04-02T00:00:00Z

provider

required

One of four hosted-inference providers.

Type: string

Example provider-a

model_family

required

Open-weight lineage the listing is grouped under (Llama, Qwen, Mixtral, DeepSeek).

Type: string

Example llama

model_listing

nullable

Provider raw model product name before family grouping.

Type: string

Example llama-3.1-8b-instruct

output_price_per_mtok_usd

required

Normalized output-token price - the quantity the study actually measures.

Type: float
Unit: USD / 1M output tokens

Example 0.02

input_price_per_mtok_usd

nullable

Input-token price where listed.

Type: float
Unit: USD / 1M input tokens

Example 0.01

model_tier

nullable

Coarse tier flag; the study does not pool tiers.

Type: string

Example small_open

context_window

nullable

Context window listed for the model at that snapshot.

Type: int
Unit: tokens

Example 131072

instance_total

nullable

Marketplace inventory: instances listed for the provider and SKU.

Type: int

Example 2368

instance_rented

nullable

Marketplace inventory: instances currently rented.

Type: int

Example 1248

utilization_ratio

nullable

Rented over total for the provider, SKU, and region at that snapshot. The supply-tightness read.

Type: float
Unit: 0..1

Example 0.527

Field	Type	Constraint	Description
observed_at	timestamp · UTC	required	When the price was collected from the provider pricing surface. e.g. 2026-04-02T00:00:00Z
provider	string	required	One of four hosted-inference providers. e.g. provider-a
model_family	string	required	Open-weight lineage the listing is grouped under (Llama, Qwen, Mixtral, DeepSeek). e.g. llama
model_listing	string	nullable	Provider raw model product name before family grouping. e.g. llama-3.1-8b-instruct
output_price_per_mtok_usd	float · USD / 1M output tokens	required	Normalized output-token price - the quantity the study actually measures. e.g. 0.02
input_price_per_mtok_usd	float · USD / 1M input tokens	nullable	Input-token price where listed. e.g. 0.01
model_tier	string	nullable	Coarse tier flag; the study does not pool tiers. e.g. small_open
context_window	int · tokens	nullable	Context window listed for the model at that snapshot. e.g. 131072
instance_total	int	nullable	Marketplace inventory: instances listed for the provider and SKU. e.g. 2368
instance_rented	int	nullable	Marketplace inventory: instances currently rented. e.g. 1248
utilization_ratio	float · 0..1	nullable	Rented over total for the provider, SKU, and region at that snapshot. The supply-tightness read. e.g. 0.527

03What's included

Token Price Index

Input and output token rates per model per provider over time - the unit economics of inference, tracked continuously.

Latency & Throughput

Time-to-first-token, generation throughput, and latency percentiles - congestion signals that move before capacity announcements.

Reliability Signal

Request success and error rates across providers - a real-time read on where inference demand is outrunning supply.

04Methodology

How it is built

01
Collection
Collect from the public pricing surfaces of four hosted-inference providers.
02
Normalization
Normalize to a per-million-token basis for output tokens.
03
Family grouping
Group heterogeneous model listings by the underlying open-weight family so comparisons stay within recognizable lineages.
04
Tier separation
Observed models span small open-weight checkpoints to frontier-scale hosted deployments priced very differently; tiers are not pooled into a single level.
05
Direction-over-level reporting
Report the direction - a steep decline at the cheap end - as robust, and treat absolute price levels as preliminary because the cheapest quote in any period may reflect a different tier than in another.

05Evals

How we validate

What each evaluation measures and how it is run. Where no benchmark is published, we show the methodology and say so.

Cheapest-output price decline

Measures

How the price of the cheapest available output tokens moved over time.

Method

Track the cheapest available output price per million tokens across the four providers over the window.

Result

Descriptive, published as preliminary: fell from about $0.13 per million tokens in mid-2024 into the $0.01 to $0.03 range in 2025-2026. Direction robust; absolute level preliminary due to mixed tiers.

Marketplace utilization as a supply-tightness read

Measures

Whether rented-versus-available instance counts identify capacity tightening on a marketplace.

Method

Poll marketplace inventory per provider, SKU, and region, recording instances total, available, and rented, and derive a utilization ratio per snapshot.

Result

Real measured: 1,366 utilization snapshots across the reference window, every row carrying a computed ratio. Latency and success-rate telemetry is a separate collection surface and is quoted only where it is actually collected.

06Graders

Ground truth

What correct means for this data, and how it is established.

Ground truth

For the descriptive layer, the observed posted output-token prices across the four providers on a cheapest-available basis. For the proposed demand signal, there is no ground truth yet because the data is uncollected.

How it is established

Descriptive aggregation of the cheapest-available output price over time per provider and family. No predictive grader exists; the proposed demand-side validation is future work gated on API access.

07Application

Demand-Side Compute Read

Throughput collapse and rising latency across providers signal surging model demand days to weeks before it shows up in chip orders.

Inference Margin Tracking

Track the falling price of intelligence per token and model the margin structure of hosted-inference and model-API businesses.

Provider Competitive Map

Compare price, speed, and reliability across providers for the same model - who is winning the inference market in real time.

08Environment & integration

How you load it

Delivery

S3, REST API, Parquet

Formats

Parquet, JSON, CSV

Auth

Licensed for internal research and model development. Sourced from public pricing surfaces; no PII or MNPI.

Cadence

Continuous polling with daily aggregates.

09Related research

The Falling Price of InferenceRead →

Request a sample