BroadRouter

Novita AI API pricing

Updated

Novita AI lists standard per-token prices on its own pricing page for 67 models, 1 of them free. The cheapest is Llama 3.1 8B Instruct at USD 0.02 input and USD 0.05 output per 1M tokens; the dearest is Kimi K3 at USD 3 / USD 15. It publishes cached-input prices for 43 of them, plus fast prices. Checked 2026-11-14.

Novita AI hosts models it does not train. By maker: Qwen 13, Z.ai 13, Moonshot AI 8, DeepSeek 7, inclusionAI 5, Meta 4, and 10 more. 68 prices, each read from Novita AI's own pricing page, checked 2026-11-14. BroadRouter is BroadNet's own site and does not sell Novita AI's services. Automated re-check 2026-11-16: 68 of 68 prices still matched the provider's page.

Models with a standard price
67
Median input / output
USD 0.385 / USD 1.375
Cached-input prices
43 of 66
Cheapest host for
6 of 58 shared models

Which models is Novita AI the cheapest provider for?

Of the 58 models that Novita AI and at least one other provider both price at the standard tier, Novita AI has the lowest price for 6, ranked on a blend of 3 input tokens to 1 output token. It also ties the lowest price for 9 more, where another provider is listed first. It is the dearest for 15. Another 8 models have a standard price from Novita AI only.

Models where Novita AI is the cheapest provider
ModelNovita AI input / outputNext cheapestSaving on 3:1 blendProviders
Qwen3.7 MaxUSD 1.25 / USD 3.75Alibaba Cloud USD 2.50 / USD 7.5050%5
GLM 4.5 AirUSD 0.13 / USD 0.85SiliconFlow USD 0.14 / USD 0.863%4
NVIDIA Nemotron 3 Nano 30B A3BUSD 0.05 / USD 0.20Amazon Bedrock USD 0.06 / USD 0.2417%3
Qwen-MT PlusUSD 0.25 / USD 0.75Alibaba Cloud USD 2.46 / USD 7.3790%2
Llama 3 8B Stheno V3.2USD 0.05 / USD 0.05Featherless USD 0.37 / USD 0.3887%2
Kimi K2 InstructUSD 0.57 / USD 2.30Featherless USD 0.60 / USD 2.507%2

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

Novita AI on the most widely hosted models
ModelNovita AIRankCheapest
GLM 5.3USD 1.40 / USD 4.406 of 24DeepInfra USD 0.5625 / USD 2.50
OpenAI GPT OSS 120BUSD 0.05 / USD 0.254 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 Novita AI sell?

Novita AI prices models from 15 makers. Qwen leads with 13 models. Standard prices, USD per 1M input / output tokens.

Novita AI models by maker
MakerModelsCheapest modelMedian input / output
Qwen13Qwen3.8 Flash USD 0.15 / USD 0.47USD 0.40 / USD 3
Z.ai13AutoGLM Phone 9B Multilingual USD 0.035 / USD 0.138USD 0.55 / USD 1.80
Moonshot AI8Kimi K2 Instruct USD 0.57 / USD 2.30USD 0.70 / USD 3.20
DeepSeek7DeepSeek V4.1 Flash USD 0.30 / USD 1.20USD 0.70 / USD 2.50
inclusionAI5Ling 3.0 Flash USD 0.06 / USD 0.18USD 0.075 / USD 0.22
Meta4Llama 3.1 8B Instruct USD 0.02 / USD 0.05USD 0.1575 / USD 0.495
Google3Gemma 3 27B USD 0.119 / USD 0.20USD 0.13 / USD 0.40
Tencent3Hy3 USD 0.14 / USD 0.58USD 0.417 / USD 1.2505
Mind Lab2Macaron V1 Tall USD 0.2475 / USD 1.43USD 0.53625 / USD 1.9525
StepFun2Step 3.7 Flash USD 0.20 / USD 1.15USD 0.60 / USD 1.925
Xiaomi2MiMo V2.6 Flash USD 0.14 / USD 0.28USD 0.2875 / USD 0.575
Microsoft1WizardLM 2 8x22B USD 0.62 / USD 0.62—
NVIDIA1NVIDIA Nemotron 3 Nano 30B A3B USD 0.05 / USD 0.20—
OpenAI1OpenAI GPT OSS 120B USD 0.05 / USD 0.25—
Sao10K1Llama 3 8B Stheno V3.2 USD 0.05 / USD 0.05—

Novita AI price list

USD per 1M tokens: 68 prices for 67 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 Novita AI is cheaper. Of 31 hosted model prices with a maker price to compare, 8 are below the maker's and 1 above.

Novita AI API prices by model
ModelTierInputCachedOutputContextvs makerSource
Apodex 1.1 MinifreestandardUSD 0—USD 0262,144—source
AutoGLM Phone 9B MultilingualstandardUSD 0.035—USD 0.13865,536—source
DeepSeek R1 0528standardUSD 0.70USD 0.35USD 2.50163,840—source
DeepSeek R1 TurbostandardUSD 0.70—USD 2.5064,000—source
DeepSeek V4 Flash 0731standardUSD 0.44USD 0.028USD 1.321,048,576—source
DeepSeek V4 ProstandardUSD 1.60USD 0.135USD 3.201,048,576+1%source
DeepSeek V4 Pro 0813standardUSD 1.32USD 0.044USD 3.961,048,576samesource
DeepSeek V4.1 FlashstandardUSD 0.30USD 0.006USD 1.201,048,576samesource
DeepSeek V4.1 Flash 1017standardUSD 0.30USD 0.006USD 1.201,048,576samesource
Gemma 3 27BstandardUSD 0.119—USD 0.2098,304—source
Gemma 4 26B A4BstandardUSD 0.13—USD 0.40262,144—source
Gemma 4 31BstandardUSD 0.14—USD 0.40262,144—source
GLM 4.5 AirstandardUSD 0.13USD 0.025USD 0.85131,072−27%source
GLM 4.5VstandardUSD 0.60USD 0.11USD 1.8065,536samesource
GLM 4.6standardUSD 0.55USD 0.11USD 2.20204,800−4%source
GLM 4.6VstandardUSD 0.30USD 0.055USD 0.90131,072samesource
GLM 4.7standardUSD 0.60USD 0.11USD 2.20204,800samesource
GLM 4.7 FlashstandardUSD 0.07USD 0.01USD 0.40200,000—source
GLM 5standardUSD 1USD 0.20USD 3.20202,800samesource
GLM 5.1standardUSD 1.38USD 0.26USD 4.40204,800−1%source
GLM 5.2standardUSD 1.40USD 0.26USD 4.401,048,576samesource
GLM 5.3standardUSD 1.40USD 0.26USD 4.401,048,576samesource
GLM 5.3 FlashstandardUSD 0.15USD 0.03USD 0.501,048,576samesource
GLM 5.3 FlashXstandardUSD 0.37USD 0.075USD 1.251,048,576samesource
Hy3standardUSD 0.14USD 0.035USD 0.58262,144—source
Hy4promostandardUSD 0.417USD 0.021USD 1.25051,000,000—source
Hy4 PreviewstandardUSD 0.834USD 0.042USD 2.5011,000,000—source
Kimi K2 InstructstandardUSD 0.57—USD 2.30131,072—source
Kimi K2 Instruct 0905standardUSD 0.60—USD 2.50262,144—source
Kimi K2 ThinkingstandardUSD 0.60USD 0.15USD 2.50262,144—source
Kimi K2.5standardUSD 0.60USD 0.10USD 3262,144—source
Kimi K2.6standardUSD 0.80USD 0.16USD 3.40262,144−15%source
Kimi K2.7 CodestandardUSD 0.95USD 0.19USD 4262,144samesource
Kimi K3standardUSD 3USD 0.30USD 151,048,576samesource
Kimi K3.1 Prostandard
fast
USD 2
USD 4
USD 0.20
USD 0.20
USD 10
USD 20
1,048,576samesource
Ling 3.0 FlashstandardUSD 0.06USD 0.012USD 0.18262,144—source
Ling 3.0 Flash FinstandardUSD 0.075USD 0.015USD 0.22262,144—source
Ling 3.0 Flash SantestandardUSD 0.075USD 0.015USD 0.22262,144—source
Ling 3.0 Flash VLstandardUSD 0.075USD 0.015USD 0.22262,144—source
Ling 3.1 Flash 1019standardUSD 0.30USD 0.06USD 0.901,000,000—source
Llama 3 8B Stheno V3.2standardUSD 0.05—USD 0.058,192—source
Llama 3.1 8B InstructstandardUSD 0.02—USD 0.0516,384—source
Llama 3.3 70BstandardUSD 0.135—USD 0.4012,288—source
Llama 4 Maverick 17B 128E Instruct FP8standardUSD 0.27—USD 0.851,048,576—source
Llama 4 Scout 17B 16E InstructstandardUSD 0.18—USD 0.59131,072—source
Macaron V1 TallstandardUSD 0.2475USD 0.044USD 1.43262,144—source
Macaron V1 VentistandardUSD 0.825USD 0.165USD 2.4751,048,576—source
MiMo V2.6 FlashstandardUSD 0.14USD 0.0028USD 0.281,048,576—source
MiMo V2.6 ProstandardUSD 0.435USD 0.0036USD 0.871,048,576—source
NVIDIA Nemotron 3 Nano 30B A3BstandardUSD 0.05—USD 0.20262,144—source
OpenAI GPT OSS 120BstandardUSD 0.05—USD 0.25131,072—source
Qwen-MT PlusstandardUSD 0.25—USD 0.7516,384−90%source
Qwen3 VL 30B A3B InstructstandardUSD 0.20—USD 0.70131,072—source
Qwen3.5 122B A10BstandardUSD 0.40—USD 3.20262,144samesource
Qwen3.5 27BstandardUSD 0.30—USD 2.40262,144samesource
Qwen3.5 35B A3BstandardUSD 0.25—USD 2262,144samesource
Qwen3.5 397B A17BstandardUSD 0.60—USD 3.60262,144samesource
Qwen3.6 27BstandardUSD 0.60—USD 3.60262,144samesource
Qwen3.6 35B A3BstandardUSD 0.248—USD 1.485262,144−34%source
Qwen3.7 MaxstandardUSD 1.25USD 0.25USD 3.751,000,000−50%source
Qwen3.8 2.4T A95BstandardUSD 2USD 0.25USD 61,000,000samesource
Qwen3.8 27BstandardUSD 0.42USD 0.085USD 31,000,000−5%source
Qwen3.8 FlashstandardUSD 0.15USD 0.016USD 0.471,000,000samesource
Qwen3.8 MaxstandardUSD 2USD 0.25USD 61,000,000samesource
Step 3.7 FlashstandardUSD 0.20USD 0.04USD 1.15262,144—source
Step 5 PreviewstandardUSD 1USD 0.10USD 2.701,000,000—source
WizardLM 2 8x22BstandardUSD 0.62—USD 0.6265,535—source
  • Cached "—" means Novita AI's page shows no cache price for that row; budget cached tokens at the input price.
  • Promo: Novita AI's page describes 1 price as promotional or discounted; they can end without notice. Each model page has the list price where the source gives one.

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

What service tiers does Novita AI publish?

  • fast (1 row). Across 1 model with both prices, the median fast price is 2× standard. Novita AI row note: “Kimi K3.1 Pro Highspeed variant. Cache write (5m) 4 per 1M.”

Cached input: 43 of 66 standard prices include a cache-read price, a median 19% of the input price. What cached input means.

What we recorded about Novita AI's pricing page: LLM rows from the Serverless Endpoints table on novita.ai/pricing; exact decimals and context sizes cross-checked against Novita's own public model list (api.novita.ai/v3/openai/models). Models present only in that API but not on the pricing page (e.g. MiniMax M3, M2.7, M2.5, CoBuddy) are excluded because their model pages 404 or show no serverless price. Novita advertises an introductory 50% batch discount for "supported models" without listing which, so no batch rows.

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

Mostly it charges the same as the maker. Of 31 standard prices for models whose maker sells its own API, 8 are below the maker's price, 1 above and 22 the same, on the 3:1 blend. The median difference is zero. The biggest saving is Qwen-MT Plus, 90% below Alibaba Cloud. The biggest markup is DeepSeek V4 Pro, 1% above DeepSeek.

BroadNet Inference and Novita AI

BroadRouter is BroadNet's own site, so BroadNet is never ranked on this page. BroadNet Inference also serves 4 models that Novita AI prices: BroadNet Gemma 4 VL 26B at USD 0.10 / USD 0.22 against Novita AI's USD 0.13 / USD 0.40 (BroadNet is cheaper on the 3:1 blend); DeepSeek V4.1 Flash at USD 0.175 / USD 0.52 against Novita AI's USD 0.30 / USD 1.20 (BroadNet is cheaper on the 3:1 blend); Qwen3.8 27B at USD 0.175 / USD 2.17 against Novita AI's USD 0.42 / USD 3 (BroadNet is cheaper on the 3:1 blend); GLM 5.3 at USD 0.78 / USD 3.46 against Novita 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 Novita AI

Questions

How much does the Novita AI API cost per token?

Novita AI's standard prices run from USD 0.02 input and USD 0.05 output per 1M tokens (Llama 3.1 8B Instruct) to USD 3 / USD 15 (Kimi K3). The median across 66 standard prices is USD 0.385 input and USD 1.375 output per 1M, which is USD 0.000000385 per input token. Input, cached input and output are billed separately.

What is the cheapest model on Novita AI?

Llama 3.1 8B Instruct, at USD 0.02 input and USD 0.05 output per 1M tokens (USD 0.0275 on a 3:1 input-to-output blend). The cheapest from other makers: Qwen3.8 Flash (Qwen) at USD 0.15 / USD 0.47; AutoGLM Phone 9B Multilingual (Z.ai) at USD 0.035 / USD 0.138; Kimi K2 Instruct (Moonshot AI) at USD 0.57 / USD 2.30. Apodex 1.1 Mini is listed free.

Is Novita AI cheaper than other providers?

It depends on the model. Of 58 models that other providers also price, Novita AI is the cheapest standard price for 6, for example Qwen3.7 Max, GLM 4.5 Air and NVIDIA Nemotron 3 Nano 30B A3B; it ties the lowest price, without beating it, for 9 more; another provider is cheaper for 43; and it is the dearest for 15. Its median rank is 3 of a median 6 providers, on a 3:1 input-to-output blend.

Does Novita AI have batch, cached or priority pricing?

Cached input: yes, for 43 of 66 standard prices, at a median 19% of the input price. Other tiers: fast on 1 row (median input 2× standard).

Is any model free on Novita AI?

Novita AI lists 1 model at 0 input and 0 output: Apodex 1.1 Mini. Free rows usually come with tight rate limits; they are not counted in the price range above.

How much does a chatbot cost per month on Novita AI?

For 2,000 requests a day, 1,500 input tokens (60% cached), 300 output, over 30 days: USD 2.70 on Llama 3.1 8B Instruct, USD 41.11 on DeepSeek V4 Flash 0731 (the middle of Novita AI's list) and USD 394.20 on Kimi K3. Try your own numbers.

Read as MarkdownAsk ChatGPTAsk ClaudeAsk Perplexity