BroadRouter

Inference.net API pricing

Updated

Inference.net lists standard per-token prices for 7 models. The cheapest is GLM 5.3 Flash at USD 0.10 input and USD 0.45 output per 1M tokens; the dearest is Kimi K3 at USD 3 / USD 15. It publishes cached-input prices for all of them, and lists no per-model batch, flex or priority prices on the pages we read. Checked 2026-11-16.

Inference.net hosts models it does not train. By maker: DeepSeek 2, Moonshot AI 2, Z.ai 2, MiniMax 1. 7 prices, each read from Inference.net's own pricing page, checked 2026-11-16. BroadRouter is BroadNet's own site and does not sell Inference.net's services.

Models with a standard price
7
Median input / output
USD 0.30 / USD 1
Cached-input prices
7 of 7
Cheapest host for
1 of 7 shared models

Which models is Inference.net the cheapest provider for?

Of the 7 models that Inference.net and at least one other provider both price at the standard tier, Inference.net has the lowest price for 1, ranked on a blend of 3 input tokens to 1 output token.

Models where Inference.net is the cheapest provider
ModelInference.net input / outputNext cheapestSaving on 3:1 blendProviders
MiniMax M3.1 FlashUSD 0.10 / USD 0.50MiniMax USD 0.15 / USD 0.6024%10

The 6 models Inference.net sells that the most other providers also sell, with Inference.net's rank among them on the 3:1 blend.

Inference.net on the most widely hosted models
ModelInference.netRankCheapest
GLM 5.3USD 0.90 / USD 35 of 31DeepInfra USD 0.5625 / USD 2.50
GLM 5.3 FlashUSD 0.10 / USD 0.456 of 29GMI Cloud USD 0.075 / USD 0.25
Kimi K3USD 3 / USD 156 of 23Chutes USD 1.85 / USD 9.50
DeepSeek V4.1 FlashUSD 0.30 / USD 19 of 23Relace USD 0.07 / USD 0.50
DeepSeek V4.1 Flash 1017USD 0.30 / USD 19 of 23Sail Research USD 0.09 / USD 0.30
Kimi K3.1 ProUSD 2 / USD 104 of 20Sail Research USD 1.50 / USD 10

Which model makers does Inference.net sell?

Inference.net prices models from 4 makers. DeepSeek leads with 2 models. Standard prices, USD per 1M input / output tokens.

Inference.net models by maker
MakerModelsCheapest modelMedian input / output
DeepSeek2DeepSeek V4.1 Flash USD 0.30 / USD 1USD 0.30 / USD 1
Moonshot AI2Kimi K3.1 Pro USD 2 / USD 10USD 2.50 / USD 12.50
Z.ai2GLM 5.3 Flash USD 0.10 / USD 0.45USD 0.50 / USD 1.725
MiniMax1MiniMax M3.1 Flash USD 0.10 / USD 0.50—

Inference.net price list

USD per 1M tokens: 7 prices for 7 models, one row per model with any other service tiers stacked under the standard price. Sources checked 2026-11-16 unless a row says otherwise. "vs maker" compares a standard price with the model maker's own API on the 3:1 blend; a minus means Inference.net is cheaper. Of 7 hosted model prices with a maker price to compare, 5 are below the maker's and 0 above.

Inference.net API prices by model
ModelTierInputCachedOutputContextvs makerSource
DeepSeek V4.1 FlashstandardUSD 0.30USD 0.007USD 11,040,000−10%source
DeepSeek V4.1 Flash 1017standardUSD 0.30USD 0.007USD 11,040,000−10%source
GLM 5.3standardUSD 0.90USD 0.15USD 31,048,576−34%source
GLM 5.3 FlashstandardUSD 0.10USD 0.03USD 0.451,048,576−21%source
Kimi K3standardUSD 3USD 0.30USD 151,048,576samesource
Kimi K3.1 ProstandardUSD 2USD 0.20USD 101,048,576samesource
MiniMax M3.1 FlashstandardUSD 0.10USD 0.005USD 0.501,000,000−24%source
  • Cached "—" means Inference.net's page shows no cache price for that row; budget cached tokens at the input price.

Report a wrong price or number (email to info@broadnet.ai, with this page's link filled in).

What service tiers does Inference.net publish?

Standard prices only: Inference.net publishes no per-model batch, flex, priority or fast price on the pages we read.

Cached input: 7 of 7 standard prices include a cache-read price, a median 10% of the input price. What cached input means.

What we recorded about Inference.net's pricing page: Prices from the model table on inference.net/models (Input, Cache read, Output per 1M tokens). Inference.net is also a gateway: its catalogue data embedded in the same page marks each model as "serverless-deployment" (served by Inference.net) or "callable-route" (routed to another provider, e.g. Anthropic, OpenAI, Google Gemini, Moonshot AI or OpenRouter). Only serverless-deployment models are listed; precision and exact context length are from that embedded catalogue data (fields quantization and maxContextSize). Inference.net's own Schematron extraction models are excluded (no other provider serves them).

Is Inference.net cheaper than the model makers' own APIs?

Mostly yes. Of 7 standard prices for models whose maker sells its own API, 5 are below the maker's price, 0 above and 2 the same, on the 3:1 blend. The median difference is 10% below the maker. The biggest saving is GLM 5.3, 34% below Z.ai.

BroadNet Inference and Inference.net

BroadRouter is BroadNet's own site, so BroadNet is never ranked on this page. BroadNet Inference also serves 2 models that Inference.net prices: DeepSeek V4.1 Flash at USD 0.175 / USD 0.52 against Inference.net's USD 0.30 / USD 1 (BroadNet is cheaper on the 3:1 blend); GLM 5.3 at USD 0.78 / USD 3.46 against Inference.net's USD 0.90 / USD 3 (Inference.net is cheaper on the 3:1 blend). BroadNet prices are from its rate card, synced 2026-11-14. BroadNet Inference pricing.

More on Inference.net

Questions

How much does the Inference.net API cost per token?

Inference.net's standard prices run from USD 0.10 input and USD 0.45 output per 1M tokens (GLM 5.3 Flash) to USD 3 / USD 15 (Kimi K3). The median across 7 standard prices is USD 0.30 input and USD 1 output per 1M, which is USD 0.0000003 per input token. Input, cached input and output are billed separately.

What is the cheapest model on Inference.net?

GLM 5.3 Flash, at USD 0.10 input and USD 0.45 output per 1M tokens (USD 0.1875 on a 3:1 input-to-output blend). The cheapest from other makers: DeepSeek V4.1 Flash (DeepSeek) at USD 0.30 / USD 1; Kimi K3.1 Pro (Moonshot AI) at USD 2 / USD 10; MiniMax M3.1 Flash (MiniMax) at USD 0.10 / USD 0.50.

Is Inference.net cheaper than other providers?

It depends on the model. Of 7 models that other providers also price, Inference.net is the cheapest standard price for 1, for example MiniMax M3.1 Flash; another provider is cheaper for 6; and it is the dearest for none. Its median rank is 6 of a median 23 providers, on a 3:1 input-to-output blend.

Does Inference.net have batch, cached or priority pricing?

Cached input: yes, for 7 of 7 standard prices, at a median 10% of the input price. No batch, flex or priority prices are listed per model on the pages we read.

How much does a chatbot cost per month on Inference.net?

For 2,000 requests a day, 1,500 input tokens (60% cached), 300 output, over 30 days: USD 12.87 on MiniMax M3.1 Flash, USD 29.18 on DeepSeek V4.1 Flash 1017 (the middle of Inference.net's list) and USD 394.20 on Kimi K3. Try your own numbers.

Read as MarkdownAsk ChatGPTAsk ClaudeAsk Perplexity