BroadRouter

Together AI API pricing

Updated

Together AI lists standard prices for 25 models, 1 of them free. The cheapest is Rnj-1 Instruct at USD 0.15 input and USD 0.15 output per 1M tokens; the dearest is Kimi K3 at USD 3 / USD 15. It publishes cached-input prices for 16 of them, and lists no per-model batch, flex or priority prices on the pages we read.

Together AI hosts models it does not train. By maker: Qwen 6, DeepSeek 3, Z.ai 3, Meta 2, Moonshot AI 2, Essential AI 1, and 8 more. 25 prices, each read from Together AI's own pricing page, checked 2026-11-14. BroadRouter is BroadNet's own site and does not sell Together AI's services. Automated re-check 2026-11-16: 25 of 25 prices still matched the provider's page.

Models with a standard price
25
Median input / output
USD 0.30 / USD 1.20
Cached-input prices
16 of 24
Cheapest host for
1 of 21 shared models

Which models is Together AI the cheapest provider for?

Of the 21 models that Together AI and at least one other provider both price at the standard tier, Together AI has the lowest price for 1, ranked on a blend of 3 input tokens to 1 output token. It also ties the lowest price for 5 more, where another provider is listed first. It is the dearest for 7. Another 3 models have a standard price from Together AI only.

Models where Together AI is the cheapest provider
ModelTogether AI input / outputNext cheapestSaving on 3:1 blendProviders
Qwen3.7 PlusUSD 0.32 / USD 1.28Alibaba Cloud USD 0.40 / USD 1.6020%2

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

Together AI on the most widely hosted models
ModelTogether AIRankCheapest
GLM 5.3USD 1.40 / USD 4.406 of 24DeepInfra USD 0.5625 / USD 2.50
OpenAI GPT OSS 120BUSD 0.15 / USD 0.6011 of 21CoreWeave Forge USD 0.03 / USD 0.17
GLM 5.3 FlashUSD 0.15 / USD 0.503 of 20GMI Cloud USD 0.075 / USD 0.25
GLM 5.2USD 1.40 / USD 4.405 of 19DeepInfra USD 0.5625 / USD 1.80
Kimi K3USD 3 / USD 155 of 18Chutes USD 1.85 / USD 9.50
DeepSeek V4.1 FlashUSD 0.30 / USD 1.205 of 16DeepInfra USD 0.14 / USD 0.42

Which model makers does Together AI sell?

Together AI prices models from 13 makers. Qwen leads with 6 models. Standard prices, USD per 1M input / output tokens.

Together AI models by maker
MakerModelsCheapest modelMedian input / output
Qwen6Qwen3.5 9B USD 0.17 / USD 0.25USD 0.31 / USD 0.875
DeepSeek3DeepSeek V4.1 Flash USD 0.30 / USD 1.20USD 0.30 / USD 1.20
Z.ai3GLM 5.3 Flash USD 0.15 / USD 0.50USD 1.40 / USD 4.40
Meta2Muse Glimmer 30B USD 0.35 / USD 1.50USD 0.80 / USD 2.875
Moonshot AI2Kimi K3.1 Pro USD 2 / USD 10USD 2.50 / USD 12.50
Essential AI1Rnj-1 Instruct USD 0.15 / USD 0.15—
MiniMax1MiniMax M3.1 Flash USD 0.15 / USD 0.60—
NVIDIA1NVIDIA Nemotron 3.5 Super VL USD 0.15 / USD 0.50—
OpenAI1OpenAI GPT OSS 120B USD 0.15 / USD 0.60—
Reflection AI1Beam 1.0 501B A23B USD 0.40 / USD 1.20—
Ricursive AI1Ripple-65B Preview USD 0.20 / USD 0.60—
Thinking Machines1Inkling 2 Small USD 0.30 / USD 1—
Together AI1Tev1 4B Experimental USD 0.042 / USD 0—

Together AI price list

USD per 1M tokens: 25 prices for 25 models, one row per model with any other service tiers stacked under the standard price. Sources checked 2026-11-14 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 Together AI is cheaper. Of 13 hosted model prices with a maker price to compare, 1 is below the maker's and 0 above.

Together AI API prices by model
ModelTierInputCachedOutputContextvs makerSource
Beam 1.0 501B A23BstandardUSD 0.40USD 0.04USD 1.20not stated—source
DeepSeek V4 Pro 0813standardUSD 1.32USD 0.13USD 3.96not statedsamesource
DeepSeek V4.1 FlashstandardUSD 0.30USD 0.006USD 1.20not statedsamesource
DeepSeek V4.1 Flash 1017standardUSD 0.30USD 0.006USD 1.20not statedsamesource
GLM 5.2standardUSD 1.40USD 0.26USD 4.40not statedsamesource
GLM 5.3standardUSD 1.40USD 0.26USD 4.40not statedsamesource
GLM 5.3 FlashstandardUSD 0.15USD 0.03USD 0.50not statedsamesource
Inkling 2 SmallstandardUSD 0.30USD 0.03USD 1not stated—source
Kimi K3standardUSD 3USD 0.30USD 15not statedsamesource
Kimi K3.1 ProstandardUSD 2USD 0.20USD 10not statedsamesource
MiniMax M3.1 FlashstandardUSD 0.15USD 0.015USD 0.60not statedsamesource
Muse Glimmer 30BstandardUSD 0.35USD 0.04USD 1.50not stated—source
Muse Spark OpenstandardUSD 1.25USD 0.15USD 4.25not stated—source
NVIDIA Nemotron 3.5 Super VLstandardUSD 0.15USD 0.02USD 0.50not stated—source
OpenAI GPT OSS 120BstandardUSD 0.15—USD 0.60not stated—source
Qwen2.5 7B Instruct TurbostandardUSD 0.30—USD 0.30not stated—source
Qwen3.5 9BstandardUSD 0.17—USD 0.25not stated—source
Qwen3.6 PlusstandardUSD 0.50—USD 3not statedsamesource
Qwen3.7 MaxstandardUSD 2.50USD 0.50USD 7.50not statedsamesource
Qwen3.7 PlusstandardUSD 0.32—USD 1.28not stated−20%source
Qwen3.8 FlashstandardUSD 0.15—USD 0.47not statedsamesource
Ripple-65B PreviewstandardUSD 0.20USD 0.02USD 0.60not stated—source
Rnj-1 InstructstandardUSD 0.15—USD 0.15not stated—source
Ternary Bonsai 27BfreestandardUSD 0—USD 0not stated—source
Tev1 4B ExperimentalstandardUSD 0.042—USD 0not stated—source
  • Cached "—" means Together AI's page shows no cache price for that row; budget cached tokens at the input price.
  • Context "not stated": Together AI's pricing page does not give context windows.

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

What service tiers does Together AI publish?

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

Cached input: 16 of 24 standard prices include a cache-read price, a median 10% of the input price. What cached input means.

What we recorded about Together AI's pricing page: Serverless Chat tab of the pricing page (the Vision tab is a subset of it; the Moderation tab listed no models). US geo-pinned prices (1.25x) are mentioned in notes, not separate rows. Together advertises Batch Inference but publishes no per-model batch rate on the pricing page, so no batch rows. Context windows are not stated on the pricing page.

Is Together AI cheaper than the model makers' own APIs?

Mostly it charges the same as the maker. Of 13 standard prices for models whose maker sells its own API, 1 is below the maker's price, 0 above and 12 the same, on the 3:1 blend. The median difference is zero. The biggest saving is Qwen3.7 Plus, 20% below Alibaba Cloud.

BroadNet Inference and Together AI

BroadRouter is BroadNet's own site, so BroadNet is never ranked on this page. BroadNet Inference also serves 2 models that Together AI prices: DeepSeek V4.1 Flash at USD 0.175 / USD 0.52 against Together AI's USD 0.30 / USD 1.20 (BroadNet is cheaper on the 3:1 blend); GLM 5.3 at USD 0.78 / USD 3.46 against Together AI's USD 1.40 / USD 4.40 (BroadNet is cheaper on the 3:1 blend). BroadNet prices are from its rate card, synced 2026-11-14. BroadNet Inference pricing.

More on Together AI

Quick answers

Is Together AI OpenAI-compatible?

Yes. Together AI's OpenAI compatibility docs list chat, completions, images, speech, transcriptions, models and the Responses API. Embeddings are only on dedicated endpoints (Together AI, OpenAI API compatibility, checked 2026-11-15).

What is Together AI's API base URL?

https://api.together.ai/v1, as Together AI documents it. US-only inference uses https://us-api.together.ai/v1 (Together AI, OpenAI API compatibility, checked 2026-11-15).

Does Together AI have a free tier?

"Together AI doesn't currently offer free trials." A minimum credit purchase of 15 USD is required to use the platform (Together AI billing docs, checked 2026-11-14). The rate-limits page gives no tier numbers for serverless models (Together AI rate limits, checked 2026-11-14).

Does Together AI have a batch API?

Yes, Batch Inference, with a completion window fixed at 24 hours. The docs list 50% off for selected models only; the discounted-models table named just openai/whisper-large-v3 when we checked, and other models run at standard rates (Together AI batch inference docs, checked 2026-11-14).

Questions

How much does the Together AI API cost per token?

Together AI's standard prices run from USD 0.15 input and USD 0.15 output per 1M tokens (Rnj-1 Instruct) to USD 3 / USD 15 (Kimi K3). The median across 24 standard prices is USD 0.30 input and USD 1.20 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 Together AI?

Rnj-1 Instruct, at USD 0.15 input and USD 0.15 output per 1M tokens (USD 0.15 on a 3:1 input-to-output blend). The cheapest from other makers: Qwen3.5 9B (Qwen) at USD 0.17 / USD 0.25; DeepSeek V4.1 Flash (DeepSeek) at USD 0.30 / USD 1.20; GLM 5.3 Flash (Z.ai) at USD 0.15 / USD 0.50. Ternary Bonsai 27B is listed free.

Is Together AI cheaper than other providers?

It depends on the model. Of 21 models that other providers also price, Together AI is the cheapest standard price for 1, for example Qwen3.7 Plus; it ties the lowest price, without beating it, for 5 more; another provider is cheaper for 15; and it is the dearest for 7. Its median rank is 3 of a median 6 providers, on a 3:1 input-to-output blend.

Does Together AI have batch, cached or priority pricing?

Cached input: yes, for 16 of 24 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 Together AI?

For 2,000 requests a day, 1,500 input tokens (60% cached), 300 output, over 30 days: USD 3.78 on Tev1 4B Experimental, USD 32.72 on DeepSeek V4.1 Flash (the middle of Together AI's list) and USD 394.20 on Kimi K3. Try your own numbers.

Read as MarkdownAsk ChatGPTAsk ClaudeAsk Perplexity