CoreWeave vs DeepInfra: LLM provider comparison

At a glance

Metric winners

There is no overall score. Each winner answers one specific question using only directly comparable data.

Fastest on shared models

CCoreWeave

1.58× typical advantage

16 exact models with complete recent speed data

Lowest token cost

DDeepInfra

$0.7713 / 1M

16 exact models using a 1K-input/500-output mix

Most models available

DDeepInfra

85 models

Complete catalog coverage across all indexed modalities

Shared-model benchmark summary

Metric	CCoreWeave☆☆☆☆☆(0)	DDeepInfra☆☆☆☆☆(0)
Speed	CoreWeave9.51 s	DeepInfra13.84 s
TTFT	CoreWeave0.35 s	DeepInfra0.70 s
TPS	CoreWeave54.5 tok/s	DeepInfra37.5 tok/s
Uptime	CoreWeave100.00%	DeepInfra99.63%
Blended	CoreWeave$1.0973 / 1M	DeepInfra$0.7713 / 1M

Model	CCoreWeave☆☆☆☆☆(0)	DDeepInfra☆☆☆☆☆(0)
ZA GLM 5.2Winner · DeepInfra	CoreWeave Speed— Worst comparable value 20.46 s TTFT— Worst comparable value 1.94 s TPS— Worst comparable value 27.0 tok/s Uptime— Best comparable value 99.96% Context 262.1K Route coreweave/fp4 Blended— Worst comparable value $2.3933 / 1M	DeepInfra Speed— Best comparable value 17.86 s TTFT— Best comparable value 1.73 s TPS— Best comparable value 31.0 tok/s Uptime— Worst comparable value 98.33% Context 1M Route deepinfra/fp4 Blended— Best comparable value $1.62 / 1M
M Kimi K2.7 CodeWinner · CoreWeave	CoreWeave Speed— Best comparable value 7.35 s TTFT— Best comparable value 0.68 s TPS— Best comparable value 75.0 tok/s Uptime— Best comparable value 100.00% Context 262.1K Route coreweave/int4 Blended— Worst comparable value $1.96 / 1M	DeepInfra Speed— Worst comparable value 11.21 s TTFT— Worst comparable value 0.79 s TPS— Worst comparable value 48.0 tok/s Uptime— Worst comparable value 98.85% Context 262.1K Route deepinfra/fp4 Blended— Best comparable value $1.66 / 1M
Q Qwen3.6 27BWinner · DeepInfra	CoreWeave Speed— Worst comparable value 18.18 s TTFT— Worst comparable value 0.33 s TPS— Worst comparable value 28.0 tok/s Uptime— Best comparable value 100.00% Context 262.1K Route coreweave/fp8 Blended— Worst comparable value $1.6 / 1M	DeepInfra Speed— Best comparable value 13.44 s TTFT— Best comparable value 0.29 s TPS— Best comparable value 38.0 tok/s Uptime— Worst comparable value 98.75% Context 262.1K Route deepinfra/fp8 Blended— Best comparable value $1.28 / 1M
D DeepSeek V4 ProWinner · DeepInfra	CoreWeave Speed— Worst comparable value 64.00 s TTFT— Worst comparable value 1.50 s TPS— Worst comparable value 8.0 tok/s Uptime— Worst comparable value 98.59% Context 1M Route coreweave/fp8 Blended— Worst comparable value $2.32 / 1M	DeepInfra Speed— Best comparable value 10.90 s TTFT— Best comparable value 0.69 s TPS— Best comparable value 49.0 tok/s Uptime— Best comparable value 99.63% Context 1M Route deepinfra/fp4 Blended— Best comparable value $1.7333 / 1M
D DeepSeek V4 FlashWinner · DeepInfra	CoreWeave Speed— Best comparable value 17.27 s TTFT— Best comparable value 0.60 s TPS— Best comparable value 30.0 tok/s Uptime— Best comparable value 99.85% Context 1M Route coreweave/fp8 Blended— Worst comparable value $0.1867 / 1M	DeepInfra Speed— Worst comparable value 21.14 s TTFT— Worst comparable value 1.14 s TPS— Worst comparable value 25.0 tok/s Uptime— Worst comparable value 98.83% Context 1M Route deepinfra/fp4 Blended— Best comparable value $0.12 / 1M
M Kimi K2.6Winner · CoreWeave	CoreWeave Speed— Best comparable value 3.30 s TTFT— Best comparable value 0.34 s TPS— Best comparable value 169.0 tok/s Uptime— Worst comparable value 99.89% Context 262.1K Route coreweave/fp4 Blended— Worst comparable value $1.9667 / 1M	DeepInfra Speed— Worst comparable value 14.24 s TTFT— Worst comparable value 0.73 s TPS— Worst comparable value 37.0 tok/s Uptime— Best comparable value 100.00% Context 262.1K Route deepinfra/fp4 Blended— Best comparable value $1.6667 / 1M
ZA GLM 5.1Winner · CoreWeave	CoreWeave Speed— Best comparable value 4.38 s TTFT— Best comparable value 0.50 s TPS— Best comparable value 129.0 tok/s Uptime— Best comparable value 100.00% Context 202.8K Route coreweave/fp8 Blended— Worst comparable value $2.4 / 1M	DeepInfra Speed— Worst comparable value 12.79 s TTFT— Worst comparable value 0.89 s TPS— Worst comparable value 42.0 tok/s Uptime— Worst comparable value 99.68% Context 202.8K Route deepinfra/fp4 Blended— Best comparable value $1.8667 / 1M
G Gemma 4 31BWinner · CoreWeave	CoreWeave Speed— Best comparable value 30.82 s TTFT— Worst comparable value 1.41 s TPS— Best comparable value 17.0 tok/s Uptime— Worst comparable value 77.67% Context 262.1K Route coreweave/bf16 Blended— Best comparable value $0.1967 / 1M	DeepInfra Speed— Worst comparable value 42.50 s TTFT— Best comparable value 0.83 s TPS— Worst comparable value 12.0 tok/s Uptime— Best comparable value 97.49% Context 262.1K Route deepinfra/turbo Blended— Worst comparable value $0.2033 / 1M
Q Qwen3.5-35B-A3BWinner · CoreWeave	CoreWeave Speed— Best comparable value 4.86 s TTFT— Worst comparable value 0.35 s TPS— Best comparable value 111.0 tok/s Uptime 100.00% Context 262.1K Route coreweave/fp8 Blended— Worst comparable value $0.5833 / 1M	DeepInfra Speed— Worst comparable value 7.98 s TTFT— Best comparable value 0.29 s TPS— Worst comparable value 65.0 tok/s Uptime 100.00% Context 262.1K Route deepinfra/fp8 Blended— Best comparable value $0.4267 / 1M
D DeepSeek V3.1Winner · CoreWeave	CoreWeave Speed— Best comparable value 9.97 s TTFT— Best comparable value 0.35 s TPS— Best comparable value 52.0 tok/s Uptime— Best comparable value 100.00% Context 161K Route coreweave/fp8 Blended— Worst comparable value $0.9167 / 1M	DeepInfra Speed— Worst comparable value 30.16 s TTFT— Worst comparable value 0.75 s TPS— Worst comparable value 17.0 tok/s Uptime— Worst comparable value 99.96% Context 163.8K Route deepinfra/fp4 Blended— Best comparable value $0.4833 / 1M
O gpt-oss-120bWinner · CoreWeave	CoreWeave Speed— Best comparable value 5.92 s TTFT— Best comparable value 0.36 s TPS— Best comparable value 90.0 tok/s Uptime— Worst comparable value 99.68% Context 131.1K Route coreweave/fp4 Blended— Best comparable value $0.0733 / 1M	DeepInfra Speed— Worst comparable value 10.39 s TTFT— Worst comparable value 0.39 s TPS— Worst comparable value 50.0 tok/s Uptime— Best comparable value 99.97% Context 131.1K Route deepinfra/bf16 Blended— Worst comparable value $0.0813 / 1M
O gpt-oss-20bWinner · CoreWeave	CoreWeave Speed— Best comparable value 4.18 s TTFT— Best comparable value 0.28 s TPS— Best comparable value 128.0 tok/s Uptime— Worst comparable value 87.61% Context 131.1K Route coreweave/fp4 Blended— Best comparable value $0.0633 / 1M	DeepInfra Speed— Worst comparable value 4.86 s TTFT— Worst comparable value 0.36 s TPS— Worst comparable value 111.0 tok/s Uptime— Best comparable value 99.98% Context 131.1K Route deepinfra/bf16 Blended— Worst comparable value $0.0667 / 1M
Q Qwen3 Coder 480B A35BWinner · DeepInfra	CoreWeave Speed— Best comparable value 9.06 s TTFT— Best comparable value 0.29 s TPS— Best comparable value 57.0 tok/s Uptime N/a Context 262.1K Route coreweave/bf16 Blended— Worst comparable value $1.1667 / 1M	DeepInfra Speed— Worst comparable value 12.74 s TTFT— Worst comparable value 0.54 s TPS— Worst comparable value 41.0 tok/s Uptime 100.00% Context 262.1K Route deepinfra/turbo Blended— Best comparable value $0.5333 / 1M
ML Llama 3.3 70B InstructWinner · DeepInfra	CoreWeave Speed— Best comparable value 18.01 s TTFT— Best comparable value 0.15 s TPS— Best comparable value 28.0 tok/s Uptime— Best comparable value 100.00% Context 128K Route coreweave/fp16 Blended— Worst comparable value $0.71 / 1M	DeepInfra Speed— Worst comparable value 63.18 s TTFT— Worst comparable value 0.68 s TPS— Worst comparable value 8.0 tok/s Uptime— Worst comparable value 97.29% Context 131.1K Route deepinfra/turbo Blended— Best comparable value $0.1733 / 1M
ML Llama 3.1 70B InstructWinner · CoreWeave	CoreWeave Speed— Best comparable value 12.50 s TTFT— Worst comparable value 0.31 s TPS— Best comparable value 41.0 tok/s Uptime— Best comparable value 100.00% Context 128K Route coreweave/bf16 Blended— Worst comparable value $0.8 / 1M	DeepInfra Speed— Worst comparable value 25.29 s TTFT— Best comparable value 0.29 s TPS— Worst comparable value 20.0 tok/s Uptime— Worst comparable value 99.73% Context 131.1K Route deepinfra/turbo Blended— Best comparable value $0.4 / 1M
ML Llama 3.1 8B InstructWinner · CoreWeave	CoreWeave Speed— Best comparable value 4.63 s TTFT— Best comparable value 0.17 s TPS— Best comparable value 112.0 tok/s Uptime— Best comparable value 100.00% Context 128K Route coreweave/bf16 Blended— Worst comparable value $0.22 / 1M	DeepInfra Speed— Worst comparable value 39.18 s TTFT— Worst comparable value 0.71 s TPS— Worst comparable value 13.0 tok/s Uptime— Worst comparable value 99.44% Context 131.1K Route deepinfra/fp8 Blended— Best comparable value $0.0267 / 1M

A per-model winner combines blended price and estimated 500-token response time with equal proportional weight. Ties and rows missing either measurement receive no badge. Comparison data calculated Jul 24, 2026, 6:19 PM UTC. Values use one deterministic route per provider and model; missing measurements remain visible as N/a.