Nebius

Inference host · 17 models · 17 priced routes · nebius.com

Models on Nebius

About Nebius

Nebius Group N.V., headquartered in Amsterdam, is a technology company that provides artificial intelligence infrastructure. The company also owns Avride and TripleTen, as well as stakes in Toloka and Clickhouse.

Across GlyphCompany profile acquisitions 8 dated events compute datacenters NBIS earnings calendar

Model	In / out per 1M	Vs headline
Qwen3 Embedding 8B bf16 · The Qwen3 Embedding model series is the latest proprietary model of the Qwen family, specifically designed for text embedding and ranking tasks.	$0.01 / FreeSame vs headline	Same	bf16 · The Qwen3 Embedding model series is the latest proprietary model of the Qwen family, specifically designed for text embedding and ranking tasks.
Nemotron 3 Nano 30B A3B fp8 · NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems.	$0.06 / $0.24Same vs headline	Same	fp8 · NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems.
Qwen3 30B A3B Instruct 2507 fp8 · Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference.	$0.1 / $0.3+5% vs headline	+5%	fp8 · Qwen3-30B-A3B-Instruct-2507 is a 30.5B-parameter mixture-of-experts language model from Qwen, with 3.3B active parameters per inference.
Gemma 3 27B fp8 · Gemma 3 introduces multimodality, supporting vision-language input and text outputs.	$0.1 / $0.3+50% vs headline	+50%	fp8 · Gemma 3 introduces multimodality, supporting vision-language input and text outputs.
DeepSeek V4 Flash 0731 fp8 · DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total.	$0.14 / $0.28+96% vs headline	+96%	fp8 · DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total.
GLM 5.3 Flash fp4 · GLM-5.3-Flash is a native multimodal model from Z.ai.	$0.15 / $0.5+389% vs headline	+389%	fp4 · GLM-5.3-Flash is a native multimodal model from Z.ai.
gpt-oss-120b fp4 · gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases.	$0.15 / $0.6+1467% vs headline	+1467%	fp4 · gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases.
Nemotron 3.5 Super VL bf16 · NVIDIA Nemotron 3.5 Super VL is an open multimodal mixture-of-experts model from NVIDIA, with 12B active parameters out of 120B total.	$0.3 / $0.9+240% vs headline	+240%	bf16 · NVIDIA Nemotron 3.5 Super VL is an open multimodal mixture-of-experts model from NVIDIA, with 12B active parameters out of 120B total.
DeepSeek V4.1 Flash fp8 · DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture.	$0.3 / $1.2+383% vs headline	+383%	fp8 · DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture.
DeepSeek V4.1 Flash 1017 fp8 · DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the deepseek-flash API model name.	$0.3 / $1.2+100% vs headline	+100%	fp8 · DeepSeek V4.1 Flash 1017 is the October 17, 2026 snapshot of DeepSeek V4.1 Flash, served by DeepSeek under the deepseek-flash API model name.
Qwen3.8 27B fp4 · Qwen3.8 27B is an open-weight dense vision-language model from Qwen.	$0.45 / $3+14% vs headline	+14%	fp4 · Qwen3.8 27B is an open-weight dense vision-language model from Qwen.
DeepSeek V4 Pro 0813 fp8 · DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model from DeepSeek.	$1.32 / $3.96+100% vs headline	+100%	fp8 · DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model from DeepSeek.
Muse Spark Open fp4 · Muse Spark Open is a coding-optimized open-weight model.	$1.25 / $4.25Same vs headline	Same	fp4 · Muse Spark Open is a coding-optimized open-weight model.
GLM 5.1 fp8 · GLM-5.1 delivers a major leap in coding capability, with particularly significant gains in handling long-horizon tasks.	$1.4 / $4.4+54% vs headline	+54%	fp8 · GLM-5.1 delivers a major leap in coding capability, with particularly significant gains in handling long-horizon tasks.
GLM 5.2 fp4 · GLM 5.2 is a large-scale reasoning model from Z.ai.	$1.4 / $4.4+131% vs headline	+131%	fp4 · GLM 5.2 is a large-scale reasoning model from Z.ai.
DeepSeek V4 Pro 0423 fp8 · DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window.	$1.75 / $3.5+108% vs headline	+108%	fp8 · DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window.
Kimi K3.1 Pro fp4 · Kimi K3.1 Pro is Moonshot AI's flagship open-weight multimodal reasoning model and a post-trained successor to Kimi K3, with a 1M-token context window.	$2 / $10+76% vs headline	+76%	fp4 · Kimi K3.1 Pro is Moonshot AI's flagship open-weight multimodal reasoning model and a post-trained successor to Kimi K3, with a 1M-token context window.