BroadRouter

Venice API pricing

Updated

Venice lists standard per-token prices on its own pricing page for 57 models. The cheapest is Qwen 2.5 7B at USD 0.05 input and USD 0.13 output per 1M tokens; the dearest is Kimi K3 at USD 3.75 / USD 18.75. It publishes cached-input prices for 39 of them, plus fast and ultrafast tiers. Checked 2026-11-15.

Venice hosts models it does not train. By maker: Qwen 12, DeepSeek 10, Z.ai 7, Google 5, Moonshot AI 5, MiniMax 4, and 8 more. 60 prices, each read from Venice's own pricing page, checked 2026-11-15. BroadRouter is BroadNet's own site and does not sell Venice's services. Automated re-check 2026-11-16: 60 of 60 prices still matched the provider's page.

Models with a standard price
57
Median input / output
USD 0.33 / USD 1.32
Cached-input prices
39 of 57
Cheapest host for
2 of 45 shared models

Which models is Venice the cheapest provider for?

Of the 45 models that Venice and at least one other provider both price at the standard tier, Venice has the lowest price for 2, ranked on a blend of 3 input tokens to 1 output token. It also ties the lowest price for 1 more, where another provider is listed first. It is the dearest for 18. Another 12 models have a standard price from Venice only.

Models where Venice is the cheapest provider
ModelVenice input / outputNext cheapestSaving on 3:1 blendProviders
MiniMax M2.5USD 0.27 / USD 0.95Featherless USD 0.295 / USD 1.2016%6
GLM 4.6USD 0.43 / USD 1.75DeepInfra USD 0.50 / USD 213%6

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

Venice on the most widely hosted models
ModelVeniceRankCheapest
GLM 5.3USD 1.75 / USD 5.5022 of 24DeepInfra USD 0.5625 / USD 2.50
OpenAI GPT OSS 120BUSD 0.07 / USD 0.306 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 3.75 / USD 18.7518 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 Venice sell?

Venice prices models from 14 makers. Qwen leads with 12 models. Standard prices, USD per 1M input / output tokens.

Venice models by maker
MakerModelsCheapest modelMedian input / output
Qwen12Qwen 2.5 7B USD 0.05 / USD 0.13USD 0.2875 / USD 1.70
DeepSeek10DeepSeek V4 Flash 0731 USD 0.13 / USD 0.26USD 0.33 / USD 1.26
Z.ai7GLM 5.3 Flash USD 0.15 / USD 0.50USD 1 / USD 3.20
Google5Gemma 3 27B USD 0.12 / USD 0.20USD 0.13 / USD 0.40
Moonshot AI5Kimi K2.5 USD 0.56 / USD 3.50USD 0.75 / USD 3.50
MiniMax4MiniMax M3.1 Flash USD 0.1875 / USD 0.75USD 0.285 / USD 1.075
Meta3Llama 3.2 3B Instruct USD 0.15 / USD 0.60USD 0.70 / USD 2.80
NVIDIA3NVIDIA Nemotron 3 Nano 30B A3B USD 0.075 / USD 0.30USD 0.20 / USD 0.50
Venice2Venice Uncensored 1.2 USD 0.20 / USD 0.90USD 0.35 / USD 1.45
Xiaomi2MiMo V2.6 Flash USD 0.175 / USD 0.35USD 0.2875 / USD 1.175
Mistral AI1Mistral Small 3.2 24B Instruct USD 0.09375 / USD 0.25—
olafangensan1GLM 4.7 Flash Heretic USD 0.07 / USD 0.40—
OpenAI1OpenAI GPT OSS 120B USD 0.07 / USD 0.30—
Tencent1Hy4 Preview USD 1.0425 / USD 3.12625—

Venice price list

USD per 1M tokens: 60 prices for 57 models, one row per model with any other service tiers stacked under the standard price. Sources checked 2026-11-15 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 Venice is cheaper. Of 21 hosted model prices with a maker price to compare, 6 are below the maker's and 9 above.

Venice API prices by model
ModelTierInputCachedOutputContextvs makerSource
DeepSeek V3.2standardUSD 0.33USD 0.16USD 0.48160,000—source
DeepSeek V4 FlashstandardUSD 0.138USD 0.028USD 0.2751,000,000—source
DeepSeek V4 Flash 0731standard
fast
USD 0.13
USD 0.35
USD 0.028
USD 0.0875
USD 0.26
USD 0.70
1,000,000—source
DeepSeek V4 ProstandardUSD 1.65USD 0.33USD 3.3011,000,000+4%source
DeepSeek V4 Pro 0813standardUSD 1.65USD 0.165USD 4.951,000,000+25%source
DeepSeek V4.1 FlashstandardUSD 0.30USD 0.0075USD 1.201,000,000samesource
DeepSeek V4.1 Flash 1017standard
ultrafast
USD 0.30
USD 3
USD 0.0075
USD 0.06
USD 1.20
USD 12
1,000,000samesource
DeepSeek V4.1 Flash 1017 AbliteratedstandardUSD 0.375USD 0.0075USD 1.501,000,000—source
DeepSeek V4.1 Flash 1017 UncensoredstandardUSD 0.33USD 0.00825USD 1.321,000,000—source
DeepSeek V4.1 Flash UncensoredstandardUSD 0.33USD 0.00825USD 1.321,000,000—source
Gemma 3 27BstandardUSD 0.12—USD 0.20198,000—source
Gemma 4 26B A4BstandardUSD 0.13USD 0.05USD 0.40256,000—source
Gemma 4 26B A4B UncensoredstandardUSD 0.19—USD 0.8864,000—source
Gemma 4 31BstandardUSD 0.12USD 0.09USD 0.36256,000—source
Gemma 4 UncensoredstandardUSD 0.1625—USD 0.50256,000—source
GLM 4.6standardUSD 0.43USD 0.08USD 1.75198,000−24%source
GLM 4.7standardUSD 0.55USD 0.11USD 2.65198,000+7%source
GLM 4.7 Flash HereticstandardUSD 0.07USD 0.035USD 0.40200,000—source
GLM 5standardUSD 1USD 0.20USD 3.20198,000samesource
GLM 5.1standardUSD 1.54USD 0.286USD 4.84200,000+10%source
GLM 5.2standardUSD 1.40USD 0.26USD 4.401,000,000samesource
GLM 5.3standardUSD 1.75USD 0.325USD 5.501,000,000+25%source
GLM 5.3 FlashstandardUSD 0.15USD 0.03USD 0.501,048,576samesource
Hy4 PreviewstandardUSD 1.0425USD 0.0525USD 3.126251,048,576—source
Kimi K2.5standardUSD 0.56USD 0.22USD 3.50256,000—source
Kimi K2.6standardUSD 0.75USD 0.16USD 3.50256,000−16%source
Kimi K2.7 CodestandardUSD 0.75USD 0.16USD 3.50256,000−16%source
Kimi K3standard
fast
USD 3.75
USD 4.50
USD 0.375
USD 0.45
USD 18.75
USD 22.50
1,000,000+25%source
Kimi K3.1 ProstandardUSD 2USD 0.20USD 101,000,000samesource
Llama 3.2 3B InstructstandardUSD 0.15—USD 0.60128,000—source
Llama 3.3 70BstandardUSD 0.70—USD 2.80128,000—source
MiMo v2.5standardUSD 0.40USD 0.08USD 21,000,000—source
MiMo V2.6 FlashstandardUSD 0.175USD 0.00375USD 0.351,000,000—source
MiniMax M2.5standardUSD 0.27USD 0.03USD 0.95198,000—source
MiniMax M2.7standardUSD 0.375USD 0.06875USD 1.50198,000—source
MiniMax M3 PreviewstandardUSD 0.30USD 0.06USD 1.20524,288—source
MiniMax M3.1 FlashstandardUSD 0.1875USD 0.01875USD 0.751,000,000+25%source
Mistral Small 3.2 24B InstructstandardUSD 0.09375—USD 0.25256,000—source
Muse Spark OpenstandardUSD 1.5625USD 0.1875USD 5.31251,048,576—source
NVIDIA Nemotron 3 Nano 30B A3BstandardUSD 0.075—USD 0.30128,000—source
NVIDIA Nemotron 3 UltrastandardUSD 0.625USD 0.1875USD 3.125256,000—source
NVIDIA Nemotron 3.5 Super VLstandardUSD 0.20USD 0.04USD 0.50262,144—source
OpenAI GPT OSS 120BstandardUSD 0.07—USD 0.30128,000—source
Qwen 2.5 7BstandardUSD 0.05—USD 0.1332,000—source
Qwen 3 Next 80bstandardUSD 0.35—USD 1.90256,000—source
Qwen3 235B A22B Instruct 2507 FP8standardUSD 0.15—USD 0.75128,000—source
Qwen3 235B A22B Thinking 2507 FP8standardUSD 0.45—USD 3.50128,000—source
Qwen3 Coder 480B A35BstandardUSD 0.35USD 0.04USD 1.50256,000—source
Qwen3 VL 235BstandardUSD 0.21USD 0.10USD 1.90128,000—source
Qwen3 VL 30B A3BstandardUSD 0.25—USD 0.90128,000—source
Qwen3.5 35B A3BstandardUSD 0.15USD 0.05USD 1256,000−47%source
Qwen3.6 27BstandardUSD 0.325—USD 3.25256,000−22%source
Qwen3.6 35B A3BstandardUSD 0.10—USD 1256,000−61%source
Qwen3.8 2.4T A95BstandardUSD 2.50USD 0.3125USD 7.50262,144+25%source
Qwen3.8 27BstandardUSD 0.45—USD 3.20262,144+1%source
Venice Role Play UncensoredstandardUSD 0.50—USD 2128,000—source
Venice Uncensored 1.2standardUSD 0.20—USD 0.90128,000—source
  • Cached "—" means Venice'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 Venice publish?

  • fast (2 rows). Across 2 models with both prices, the median fast price is 1.95× standard (range 1.2× standard to 2.69× standard). Venice row note: “Fast variant, model ID kimi-k3-fast-api.”
  • ultrafast (1 row). Across 1 model with both prices, the median ultrafast price is 10× standard. Venice row note: “Ultrafast variant, model ID deepseek-v4-1-flash-1017-ultrafast.”

Cached input: 39 of 57 standard prices include a cache-read price, a median 19% of the input price. What cached input means.

What we recorded about Venice's pricing page: Prices from the "Text models" table on Venice's API pricing page (docs.venice.ai/pricing, also served as pricing.md), cross-checked against model_spec.pricing in Venice's public GET https://api.venice.ai/api/v1/models, which also gives context length and quantization. Included: models on Venice's "Private" tier (Venice: "zero data retention, self-hosted open-source models") and E2EE-only models. Excluded: "Anonymized" models (Venice: "third-party models with identifying metadata stripped", i.e. resold third-party APIs such as Claude, GPT and Gemini), xAI Grok models (closed models, so not self-hosted even though labelled Private), and E2EE duplicates of a Private model (E2EE versions of GLM 5.2, GLM 5.3, GLM 5.3 Flash, GPT OSS 120B, Kimi K2.6, Kimi K3, Qwen 3.6 35B A3B, Qwen 3.8 27B and DeepSeek V4 Flash cost slightly more). No long-context surcharge applies to the included models.

Is Venice cheaper than the model makers' own APIs?

Mostly no. Of 21 standard prices for models whose maker sells its own API, 6 are below the maker's price, 9 above and 6 the same, on the 3:1 blend. The median difference is zero. The biggest saving is Qwen3.6 35B A3B, 61% below Alibaba Cloud. The biggest markup is MiniMax M3.1 Flash, 25% above MiniMax.

BroadNet Inference and Venice

BroadRouter is BroadNet's own site, so BroadNet is never ranked on this page. BroadNet Inference also serves 4 models that Venice prices: BroadNet Gemma 4 VL 26B at USD 0.10 / USD 0.22 against Venice'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 Venice'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 Venice's USD 0.45 / USD 3.20 (BroadNet is cheaper on the 3:1 blend); GLM 5.3 at USD 0.78 / USD 3.46 against Venice's USD 1.75 / USD 5.50 (BroadNet is cheaper on the 3:1 blend). BroadNet prices are from its rate card, synced 2026-11-14. BroadNet Inference pricing.

More on Venice

Questions

How much does the Venice API cost per token?

Venice's standard prices run from USD 0.05 input and USD 0.13 output per 1M tokens (Qwen 2.5 7B) to USD 3.75 / USD 18.75 (Kimi K3). The median across 57 standard prices is USD 0.33 input and USD 1.32 output per 1M, which is USD 0.00000033 per input token. Input, cached input and output are billed separately.

What is the cheapest model on Venice?

Qwen 2.5 7B, at USD 0.05 input and USD 0.13 output per 1M tokens (USD 0.07 on a 3:1 input-to-output blend). The cheapest from other makers: DeepSeek V4 Flash 0731 (DeepSeek) at USD 0.13 / USD 0.26; GLM 5.3 Flash (Z.ai) at USD 0.15 / USD 0.50; Gemma 3 27B (Google) at USD 0.12 / USD 0.20.

Is Venice cheaper than other providers?

It depends on the model. Of 45 models that other providers also price, Venice is the cheapest standard price for 2, for example MiniMax M2.5 and GLM 4.6; it ties the lowest price, without beating it, for 1 more; another provider is cheaper for 42; and it is the dearest for 18. Its median rank is 4 of a median 9 providers, on a 3:1 input-to-output blend.

Does Venice have batch, cached or priority pricing?

Cached input: yes, for 39 of 57 standard prices, at a median 19% of the input price. Other tiers: fast on 2 rows (median input 1.95× standard); ultrafast on 1 row (median input 10× standard).

How much does a chatbot cost per month on Venice?

For 2,000 requests a day, 1,500 input tokens (60% cached), 300 output, over 30 days: USD 6.84 on Qwen 2.5 7B, USD 38.70 on Qwen3 VL 30B A3B (the middle of Venice's list) and USD 492.75 on Kimi K3. Try your own numbers.

Read as MarkdownAsk ChatGPTAsk ClaudeAsk Perplexity