What models does CoreWeave offer?

CoreWeave offers 30 models that we track: GLM-5.3-Flash, DeepSeek V4.1 Flash (max), DeepSeek V4 Flash 0731 (max), GLM-5.2 (max), Qwen3.8 27B (xhigh) (FP8), DeepSeek V4 Pro (max), Beam 1.0 (NVFP4), MiniMax-M3 (NVFP4), Kimi K2.6, Nemotron 3.5 Super VL (NVFP4), Kimi K2.7 Code, DeepSeek V4.1 Flash (non-reasoning), Kimi K2.6 (non-reasoning), Nemotron 3 Ultra, DeepSeek V4 Pro (non-reasoning), Qwen3.6 35B A3B (FP8), Gemma 4 26B A4B, Qwen3.6 35B A3B (non-reasoning) (FP8), Gemma 4 31B, Gemma 4 31B (non-reasoning), DeepSeek V3.1 (non-reasoning), Gemma 4 26B A4B (non-reasoning), Nemotron 3.5 Lightning (BF16), gpt-oss-120b (high), Granite 4.2 8B, gpt-oss-120b (low), gpt-oss-20b (low), gpt-oss-20b (high), Llama 3.3 70B, and Llama 3.1 8B.

Artificial Analysis

CoreWeave: Models Intelligence, Performance & Price

Name: Intelligence
Creator: Artificial Analysis
License: https://artificialanalysis.ai/docs/legal/Terms-of-Use.pdf

CoreWeave

This analysis is intended to support you in choosing the best model provided by CoreWeave for your use-case.

Most Intelligent

GLM-5.3-Flash

DeepSeek V4.1 Flash (max)

DeepSeek V4 Flash 0731 (max)

GLM-5.2 (max)

Qwen3.8 27B (xhigh) (FP8)

Intelligence Index

Total 30 models

Fastest

Nemotron 3.5 Super VL (NVFP4)

478 t/s

Nemotron 3.5 Lightning (BF16)

387 t/s

Nemotron 3 Ultra

381 t/s

GLM-5.2 (max)

327 t/s

Beam 1.0 (NVFP4)

303 t/s

Output speed

Total 30 models

Lowest Price

gpt-oss-20b (low)

$0.04

gpt-oss-20b (high)

$0.04

gpt-oss-120b (high)

$0.04

gpt-oss-120b (low)

$0.04

Nemotron 3.5 Lightning (BF16)

$0.06

Blended price (per 1M tokens)

Total 30 models

CoreWeave offers 30 models, each with different intelligence, performance, and pricing characteristics. Below is a comparison of the key metrics across models.

For intelligence, the top models on CoreWeave are GLM-5.3-Flash (42), DeepSeek V4.1 Flash (max) (39), and DeepSeek V4 Flash 0731 (max) (34).
For output speed, the fastest models are Nemotron 3.5 Super VL (NVFP4) (478 t/s), Nemotron 3.5 Lightning (BF16) (387 t/s), and Nemotron 3 Ultra (381 t/s). Speed varies significantly across models, with a 58% difference between the fastest and slowest.
For latency, Llama 3.1 8B (0.66s), Gemma 4 26B A4B (non-reasoning) (0.76s), and Llama 3.3 70B (0.82s) offer the lowest time to first answer token.
For pricing, gpt-oss-20b (low) ($0.04), gpt-oss-20b (high) ($0.04), and gpt-oss-120b (high) ($0.04) offer the lowest blended prices per 1M tokens.
For context window size, Beam 1.0 (NVFP4) (1M), GLM-5.3-Flash (1M), and DeepSeek V4.1 Flash (max) (1M) support the largest context windows on CoreWeave.

Intelligence

Artificial Analysis Intelligence Index · Higher is better

Speed

Output tokens per second · Higher is better

Price

USD per 1M tokens (blended) · Lower is better

Intelligence Evaluations

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v4.3.2 incorporates 10 evaluations: AA-Briefcase v1.1, GDPval-AA v2.1, AutomationBench-AA, Terminal-Bench 4.0, SciCode, Humanity's Last Exam, GDP.pdf, CritPt, AA-Omniscience, AA-LCR v1.1

Artificial Analysis Intelligence Index v4.3.2 includes: AA-Briefcase v1.1, GDPval-AA v2.1, AutomationBench-AA, Terminal-Bench 4.0, SciCode, Humanity's Last Exam, GDP.pdf, CritPt, AA-Omniscience, AA-LCR v1.1. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.

Evaluations

Intelligence evaluations measured independently by Artificial Analysis · Higher is better

AA-Briefcase v1.1Updated

Agentic knowledge work, (Elo-500)/2000

AA-Omniscience Accuracy

Knowledge

AA-Omniscience Non-Hallucination Rate

1 - hallucination rate

AutomationBench-AA

Agentic SaaS workflows

GDPval-AA v2.1Updated

Agentic real-world work tasks, (Elo-500)/2000

Humanity's Last Exam

Reasoning & knowledge

GDP.pdf

Professional document reasoning, All-pass

Terminal-Bench 4.0

Agentic coding & terminal use

AA-LCR v1.1

Long context reasoning

SciCodeUnder review

Coding

CritPtUnder review

Physics reasoning

Terminal-Bench-Science 0.1New

Agentic scientific research workflows in a terminal

FrontierCode 1.1New

Agentic software engineering in open-source repositories

AA-AnalystAgent

Quantitative analysis on spreadsheets & documents

Harvey LAB-AA v1.1Updated

Legal agentic work, Hallucination-Gated All-Pass Rate

EnterpriseOps-Gym-AA

Agentic business operations

ITBench-AA

Kubernetes incident root-cause analysis

MLCR-AA

Medical long context reasoning

MMMU-Pro

Visual reasoning

While model intelligence generally translates across use cases, specific evaluations may be more relevant for certain use cases.

Intelligence Index vs. Price

Blended at 7:2:1 (cache-input-output) · USD per 1M tokens (blended)

Most attractive quadrant

Pareto line

While higher intelligence models are typically more expensive, they do not all follow the same price-quality curve.

Context Window

Context window: tokens limit · Higher is better

Larger context windows are relevant to RAG (Retrieval Augmented Generation) LLM workflows which typically involve reasoning and information retrieval of large amounts of data.

Maximum number of combined input & output tokens. Output tokens commonly have a significantly lower limit (varied by model).

Pricing

Intelligence Index vs. Price

Blended at 7:2:1 (cache-input-output) · USD per 1M tokens (blended)

Most attractive quadrant

Pareto line

While higher intelligence models are typically more expensive, they do not all follow the same price-quality curve.

Performance Summary

Output Speed vs. Price

Output speed: output tokens per second · USD per 1M tokens (blended)

Most attractive quadrant

Pareto line

Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).

Price per token, shown in USD per million tokens. Price is a blend of cache hit, input, and output token prices using the selected ratio (default 7:2:1 cache-input-output).

Speed

Measured by Output Speed (tokens per second)

Output Speed

Output tokens per second · Higher is better

Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).

Figures represent performance of the model's first-party API or the median across providers where a first-party API is not available.

Latency

Measured by Time (seconds) to First Token

Latency: Time To First Answer Token

Seconds to first answer token received · Accounts for reasoning model 'thinking' time

Time to first answer token received, in seconds, after API request sent. For reasoning models, this includes the 'thinking' time of the model before providing an answer. For models which do not support streaming, this represents time to receive the completion.

Cache Behaviour

Cache Hit Rate

Share of cacheable input tokens served from cache · Median of the last four weeks, updated Sep 26, 2026

The share of cacheable input tokens served from the provider's prompt cache, measured during multi-turn agentic coding sessions. We regularly measure each endpoint and report a representative rate based on recent measurements. Higher values mean more of the input is billed at the cheaper cache-hit price, where the provider offers one, which lowers the effective cost of agentic workloads. For further details, see our methodology page.

Cost per Task vs. Cache Hit Rate

Weighted average cost (USD) per Intelligence Index task · Cache hit rate: median of the last four weeks, updated Sep 26, 2026

Most attractive quadrant

Pareto line

Weighted average cost per Intelligence Index task. Each evaluation’s cost is calculated from input, cache hit, cache write, reasoning, and answer token prices, divided by task count, and weighted by its Intelligence Index weight.

End-to-End Response Time

Seconds to output 500 tokens, calculated based on time to first token, 'thinking' time for reasoning models, and output speed

End-to-End Response Time vs. Price

End-to-end response time: end-to-end seconds to output 500 tokens · USD per 1M tokens (blended)

Most attractive quadrant

Pareto line

Price per token, shown in USD per million tokens. Price is a blend of cache hit, input, and output token prices using the selected ratio (default 7:2:1 cache-input-output).


									Further Analysis
GLM-5.3-Flash	1M	Open	42	$0.65	271	1.18	10.42	7.39	Model Provider
DeepSeek V4.1 Flash (max)	1M	Open	39	$1.03	224	0.88	12.05	8.94	Model Provider
DeepSeek V4 Flash 0731 (max)	262k	Open	34	$0.42	114	1.56	23.48	17.54	Model Provider
GLM-5.2 (max)	262k	Open	34	$0.70	327	1.07	8.72	6.12	Model Provider
Qwen3.8 27B (xhigh) (FP8)	262k	Open	34	$0.92	67	1.71	38.97	29.81	Model Provider
DeepSeek V4 Pro (max)	1M	Open	30	$2.00	130	1.29	38.92	33.77	Model Provider
MiniMax-M3 (NVFP4)	262k	Open	29	--	122	1.24	21.66	16.33	Model Provider
GLM-5 (FP8)	200k	Open	28*	--	--	--	--	--	Model Provider
Kimi K2.6	262k	Open	27	$1.11	134	1.09	38.04	33.22	Model Provider
Nemotron 3.5 Super VL (NVFP4)	1M	Open	26	$0.24	456	1.17	6.66	4.39	Model Provider
Kimi K2.7 Code	262k	Open	26	$0.83	205	1.08	14.40	10.88	Model Provider
DeepSeek V4.1 Flash (non-reasoning)	1M	Open	25	$0.70	251	0.91	2.90	--	Model Provider
Kimi K2.6 (non-reasoning)	262k	Open	24*	--	87	1.11	6.86	--	Model Provider
Nemotron 3 Ultra	262k	Open	23	$0.51	381	0.93	8.21	5.97	Model Provider
DeepSeek V4 Pro (non-reasoning)	1M	Open	21*	--	132	1.44	5.24	--	Model Provider
Qwen3.6 35B A3B (FP8)	262k	Open	18	$0.31	167	0.98	36.20	32.23	Model Provider
Gemma 4 26B A4B	262k	Open	17*	--	72	0.66	35.17	27.61	Model Provider
Qwen3.6 35B A3B (non-reasoning) (FP8)	262k	Open	15*	--	170	1.01	3.95	--	Model Provider
Gemma 4 31B	262k	Open	15	$0.07	58	2.00	40.74	30.08	Model Provider
Gemma 4 31B (non-reasoning)	262k	Open	14*	--	40	1.97	14.34	--	Model Provider
DeepSeek V3.1 (non-reasoning)	128k	Open	14*	--	94	1.11	6.44	--	Model Provider
Gemma 4 26B A4B (non-reasoning)	262k	Open	13*	--	78	0.76	7.20	--	Model Provider
Nemotron 3.5 Lightning (BF16)	262k	Open	13	$0.07	394	0.57	6.92	5.08	Model Provider
gpt-oss-120b (high)	131k	Open	12	$0.02	54	1.21	47.42	36.97	Model Provider
Granite 4.2 8B	131k	Open	11	$0.05	88	0.72	29.07	22.68	Model Provider
gpt-oss-120b (low)	131k	Open	10*	--	49	1.21	52.65	41.15	Model Provider
gpt-oss-20b (low)	131k	Open	10*	--	171	0.56	15.18	11.70	Model Provider
gpt-oss-20b (high)	131k	Open	9	$0.01	170	0.55	15.27	11.78	Model Provider
Llama 3.3 70B	128k	Open	8*	--	93	0.82	6.20	--	Model Provider
Llama 3.1 8B	128k	Open	7*	--	142	0.66	4.18	--	Model Provider

Key definitions

Maximum number of combined input & output tokens. Output tokens commonly have a significantly lower limit (varied by model).

Frequently Asked Questions

Common questions about CoreWeave

The most intelligent model available on CoreWeave is GLM-5.3-Flash with an Intelligence Index score of 42.

The fastest model on CoreWeave by output speed is Nemotron 3.5 Super VL (NVFP4) at 477.7 tokens per second.

The model with the lowest time to first answer token on CoreWeave is Llama 3.1 8B at 0.66s. Lower latency means faster initial response time.

The most affordable model on CoreWeave by blended price is gpt-oss-20b (low) at $0.04 per 1M tokens (7:2:1 cache hit/input/output ratio).

Prices on CoreWeave vary up to 18x across models, from $0.04 per 1M tokens for gpt-oss-20b (low) to $0.71 per 1M tokens for Llama 3.3 70B.

Yes, CoreWeave offers an OpenAI-compatible API, making it easy to switch from OpenAI or use existing OpenAI SDK integrations.

Yes, all 30 models on CoreWeave support JSON mode for structured output.

Yes, all 30 models on CoreWeave support function calling (tool use).

Yes, CoreWeave offers 21 reasoning models: GLM-5.3-Flash, DeepSeek V4.1 Flash (max), DeepSeek V4 Flash 0731 (max), GLM-5.2 (max), Qwen3.8 27B (xhigh) (FP8), DeepSeek V4 Pro (max), Beam 1.0 (NVFP4), MiniMax-M3 (NVFP4), Kimi K2.6, Nemotron 3.5 Super VL (NVFP4), Kimi K2.7 Code, Nemotron 3 Ultra, Qwen3.6 35B A3B (FP8), Gemma 4 26B A4B, Gemma 4 31B, Nemotron 3.5 Lightning (BF16), gpt-oss-120b (high), Granite 4.2 8B, gpt-oss-120b (low), gpt-oss-20b (low), and gpt-oss-20b (high). Reasoning models use extended thinking to work through complex problems before providing an answer.

Yes, all 30 models on CoreWeave are open weight models.

Yes, provider performance can vary over time due to infrastructure changes, load balancing, and updates. We continuously benchmark all providers and display historical performance trends in the "Over Time" charts.

When choosing a model on CoreWeave, consider: intelligence (for quality-sensitive tasks), output speed (for throughput-intensive tasks), latency (for interactive applications requiring quick first responses), pricing (for cost-sensitive workloads), and features like context window size, JSON mode, or function calling support.

CoreWeave: Models Intelligence, Performance & Price

Most Intelligent

Fastest

Lowest Price

Intelligence