CoreWeave

Inference host · 20 models · 20 priced routes · coreweave.com

Models on CoreWeave

About CoreWeave

CoreWeave, Inc. is an American artificial intelligence (AI) cloud-computing company based in Livingston, New Jersey. It specializes in providing cloud-based graphics processing unit (GPU) infrastructure to AI developers and enterprises, and also develops its own chip management software.

Across GlyphCompany profile 2 acquisitions Next earnings 2026-11-09 2 conferences

Model	In / out per 1M	Vs headline
gpt-oss-20b fp4 · gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license.	$0.03 / $0.13+659% vs headline	+659%	fp4 · gpt-oss-20b is an open-weight 21B parameter model released by OpenAI under the Apache 2.0 license.
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.	$0.03 / $0.17+288% vs headline	+288%	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 Lightning bf16 · NVIDIA Nemotron 3.5 Lightning is an open mixture-of-experts model from NVIDIA, with 3B active parameters out of 30B total.	$0.07 / $0.2+110% vs headline	+110%	bf16 · NVIDIA Nemotron 3.5 Lightning is an open mixture-of-experts model from NVIDIA, with 3B active parameters out of 30B total.
Gemma 4 26B A4B bf16 · Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind.	$0.1 / $0.3+76% vs headline	+76%	bf16 · Gemma 4 26B A4B IT is an instruction-tuned Mixture-of-Experts (MoE) model from Google DeepMind.
Gemma 4 31B fp4 · Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output.	$0.1 / $0.34Same vs headline	Same	fp4 · Gemma 4 31B Instruct is Google DeepMind's 30.7B dense multimodal model supporting text and image input with text output.
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.	$0.13 / $0.27+85% vs headline	+85%	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 nvfp4 · GLM-5.3-Flash is a native multimodal model from Z.ai.	$0.12 / $0.38+281% vs headline	+281%	nvfp4 · GLM-5.3-Flash is a native multimodal model from Z.ai.
Llama 3.1 8B Instruct bf16 · Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors.	$0.22 / $0.22+780% vs headline	+780%	bf16 · Meta's latest class of model (Llama 3.1) launched with a variety of sizes & flavors.
Nemotron 3.5 Super VL nvfp4 · NVIDIA Nemotron 3.5 Super VL is an open multimodal mixture-of-experts model from NVIDIA, with 12B active parameters out of 120B total.	$0.15 / $0.5+79% vs headline	+79%	nvfp4 · NVIDIA Nemotron 3.5 Super VL is an open multimodal mixture-of-experts model from NVIDIA, with 12B active parameters out of 120B total.
Qwen3.6 35B A3B fp8 · Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per token.	$0.1 / $0.89+520% vs headline	+520%	fp8 · Qwen3.6-35B-A3B is an open-weight multimodal model from Alibaba Cloud with 35 billion total parameters and 3 billion active parameters per token.
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.	$0.2 / $0.65+187% vs headline	+187%	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.
Llama 3.3 70B Instruct fp16 · The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out).	$0.71 / $0.71+358% vs headline	+358%	fp16 · The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out).
DeepSeek V3.1 fp8 · DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates.	$0.55 / $1.65+94% vs headline	+94%	fp8 · DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates.
Qwen3.8 27B fp8 · Qwen3.8 27B is an open-weight dense vision-language model from Qwen.	$0.4 / $3+10% vs headline	+10%	fp8 · Qwen3.8 27B is an open-weight dense vision-language model from Qwen.
GLM 5.2 fp4 · GLM 5.2 is a large-scale reasoning model from Z.ai.	$0.62 / $2.42+15% vs headline	+15%	fp4 · GLM 5.2 is a large-scale reasoning model from Z.ai.
Kimi K2.7 Code int4 · MoonshotAI: Kimi K2.7 Code is a coding-focused model in Moonshot AI's Kimi K2 family, built to complete end-to-end programming tasks reliably over long contexts.	$0.6 / $3-30% vs headline	-30%	int4 · MoonshotAI: Kimi K2.7 Code is a coding-focused model in Moonshot AI's Kimi K2 family, built to complete end-to-end programming tasks reliably over long contexts.
GLM 5.3 nvfp4 · GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks.	$0.75 / $3+171% vs headline	+171%	nvfp4 · GLM-5.3 is a large-scale reasoning model from Z.ai, built for complex software engineering and long-horizon agent tasks.
Kimi K2.6 fp4 · Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration.	$0.65 / $3.41-22% vs headline	-22%	fp4 · Kimi K2.6 is Moonshot AI's next-generation multimodal model, designed for long-horizon coding, coding-driven UI/UX generation, and multi-agent orchestration.
DeepSeek V4 Pro 0813 fp8 · DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model from DeepSeek.	$1.31 / $3.96+99% vs headline	+99%	fp8 · DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model from DeepSeek.