Nebius vs Together: 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

NNebius

1.43× typical advantage

3 exact models with complete recent speed data

Lowest token cost

NNebius

$2.5067 / 1M

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

Most models available

TTogether

23 models

Complete catalog coverage across all indexed modalities

Shared-model benchmark summary

Metric	NNebius☆☆☆☆☆(0)	TTogether☆☆☆☆☆(0)
Speed	Nebius11.76 s	Together16.85 s
TTFT	Nebius0.83 s	Together1.75 s
TPS	Nebius46.0 tok/s	Together37.0 tok/s
Uptime	Nebius94.36%	Together94.97%
Blended	Nebius$2.5067 / 1M	Together$2.78 / 1M

Model	NNebius☆☆☆☆☆(0)	TTogether☆☆☆☆☆(0)
M Kimi K3Winner · Nebius	Nebius Speed— Best comparable value 11.76 s TTFT— Best comparable value 0.89 s TPS— Best comparable value 46.0 tok/s Uptime— Worst comparable value 94.74% Context 8K Route nebius/fp4 Blended $7 / 1M	Together Speed— Worst comparable value 16.85 s TTFT— Worst comparable value 3.34 s TPS— Worst comparable value 37.0 tok/s Uptime— Best comparable value 94.97% Context 1M Route together Blended $7 / 1M
O gpt-oss-120bWinner · Nebius	Nebius Speed— Best comparable value 5.14 s TTFT— Best comparable value 0.42 s TPS— Best comparable value 106.0 tok/s Uptime— Best comparable value 94.36% Context 131.1K Route nebius/fp4 Blended $0.3 / 1M	Together Speed— Worst comparable value 15.27 s TTFT— Worst comparable value 1.75 s TPS— Worst comparable value 37.0 tok/s Uptime— Worst comparable value 63.29% Context 131.1K Route together Blended $0.3 / 1M
ML Llama 3.3 70B InstructWinner · Nebius	Nebius Speed— Worst comparable value 39.29 s TTFT— Worst comparable value 0.83 s TPS— Worst comparable value 13.0 tok/s Uptime— Worst comparable value 57.28% Context 131.1K Route nebius/fp8 Blended— Best comparable value $0.22 / 1M	Together Speed— Best comparable value 27.04 s TTFT— Best comparable value 0.72 s TPS— Best comparable value 19.0 tok/s Uptime— Best comparable value 96.04% Context 131.1K Route together/fp8 Blended— Worst comparable value $1.04 / 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 28, 2026, 2:19 AM UTC. Values use one deterministic route per provider and model; missing measurements remain visible as N/a.