Agent
Sign in

Inference provider · Fireworks AI

Fireworks AI

An inference provider listed on OpenRouter. Its prices here are the ones OpenRouter’s public API reported for its endpoints.

Prices reported by OpenRouter’s public API, snapshot · third-party-reported, not measured by Agent

Where its price sits

0 of 2models where its price is the cheapest, ties included

Models with two or more standard-tier providers. Ties included in “cheapest”; a calculation on reported prices, 3 input : 1 output tokens. Not a score.

At a glance

2

models in the index (of 27)

Listed under Fireworks

2.4x

median price vs the cheapest provider of the same model

Calculation · n = 2 models with 2 or more providers

99%

median uptime over the last day, as reported

n = 2 endpoints; reported, not measured

Not reported

latency and throughput

The keyless API returned latency for 0 of 265 endpoints. Gateway delay is not measured: no key in the environment.

Its price vs every other provider

Its cheapest standard-tier endpoint per model, against every other provider of that model. Switch the price type; select another provider to pin it instead. Blended price, position and “vs cheapest” are calculations on the reported prices. A cheaper endpoint can run lower precision or a shorter context.

Fireworks AI: its price against every other provider

Reported
Calculation: (3 × input + output) ÷ 4

Pinned Fireworks: on 2 of 2 rows

Model
  1. Kimi K319 providers · Relace $3.87 → Alibaba $6.90
  2. GLM 5.332 providers · Novita $0.645 → Relace $3.02
  • one provider (blended: a calculation, hollow)
  • first-party list price

Third-party reported values. 51 listings, 4 series: Blended 3:1, Input, Output, Cache read. Blended 3:1: highest Kimi K3 · Alibaba $6.90 (n 19). Lowest GLM 5.3 · Novita (fp8) $0.65 (n 32). Input: highest Kimi K3 · Alibaba $3.45 (n 19). Lowest GLM 5.3 · Relace $0.03 (n 32).

Notesn 19–32 per row

One row per model it serves; Fireworks AI is pinned in color, every other provider is gray. USD per million tokens, snapshot 2026-10-06.

Prices reported by OpenRouter’s public API (third-party-reported, not measured by Agent). One dot per provider: its cheapest standard-tier endpoint. The diamond is the first-party list price. Spread: highest ÷ lowest price of the row. A parenthesis names the quantization the provider reported. n = providers with a standard-tier endpoint.

Source: OpenRouter public API: models and provider endpoints (snapshot)

Price per provider: Kimi K3

Reported by OpenRouter’s public API, snapshot October 6, 2026. Third-party-reported prices, not measured by Agent.

Reported

19 providers with a standard-tier endpoint · cheapest Relace at $3.87 blended, priciest Alibaba at $6.90 (1.8x) · showing 19 of 19 rows

  1. InputRelace $0.83 → Alibaba $3.45 · 4.2x

  2. OutputPhala $9.75 → Alibaba $17.25 · 1.8x

  3. Cache readPhala $0.195 → AkashML $1.30 · 6.7x

  • one provider (reported price, USD per million tokens, log scale per strip)

Blended price and “vs cheapest” are calculations on the reported prices (3 input : 1 output tokens), against the cheapest standard-tier provider. A lower price can come with lower precision (fp4, fp8) or a shorter context. Uptime is what the API reported for the last day; the API reported no latency or throughput.

Every value from the study

Each row is one of its prices on its own track, with the other providers of the same chart as muted dots (6 values). Reported prices, not measured. Use Table for the plain values; each row links to the chart it copies.

Inference provider index: 27 models, 52 providers

6 values · open the study

Kimi K3 · Input
$3.00
Kimi K3 · Output
$15.00
Kimi K3 · Cache read
$0.30
GLM 5.3 · Input
$1.40
GLM 5.3 · Output
$4.40
GLM 5.3 · Cache read
$0.26

Reported by OpenRouter’s public API, not measured by AgentMuted dots: the other providers on the same chartUSD per million tokens

Fireworks AI in Inference provider index: 27 models, 52 providers: 6 values, first Kimi K3: price per million tokens by provider (Input) $3.00.

Compare Fireworks AI

Price rows state the gap and name no winner: a reported price has no interval.

Next

Write-ups that use this data

Turn the numbers into shipped work.

Agent runs these choices for you: a persistent AI worker with memory and rules, on your Claude and Codex subscriptions.

Try it on a past pull requestSee how Agent works