Nebius

Data handling can vary by model and deployment. The privacy badge above shows the more conservative of the provider-wide fallback and the policies for routes on this page; each model page shows the policy for that specific route.

Models served by Nebius

Model name

16 models

Model	Context	Input /M	Output /M	Cache read /M	Cache hit	TTFT	Throughput
DeepSeek V4 Pro 0813 FP8 deepseek/deepseek-v4-pro-0813:nebius	979k	$1.39/M	$4.16/M	$0.042/M	N/A	0.8s	112 tps
DeepSeek V4 Pro 0813 (Thinking) FP8 deepseek/deepseek-v4-pro-0813:thinking:nebius	979k	$1.39/M	$4.16/M	$0.042/M	N/A	0.8s	112 tps
DeepSeek V4.1 Flash FP8 deepseek/deepseek-v4.1-flash:nebius	1.0M	$0.315/M	$1.26/M	$0.0105/M	N/A	0.7s	83 tps
DeepSeek V4.1 Flash (Thinking) FP8 deepseek/deepseek-v4.1-flash:thinking:nebius	1.0M	$0.315/M	$1.26/M	$0.0105/M	N/A	0.7s	83 tps
DeepSeek V4.1 Flash 1017 FP8 deepseek/deepseek-v4.1-flash-1017:nebius	1.0M	$0.315/M	$1.26/M	$0.0063/M	N/A	1.6s	102 tps
DeepSeek V4.1 Flash 1017 (Thinking) FP8 deepseek/deepseek-v4.1-flash-1017:thinking:nebius	1.0M	$0.315/M	$1.26/M	$0.0063/M	N/A	1.6s	102 tps
GLM 5.1 FP8 z-ai/glm-5.1:nebius	203k	$1.47/M	$4.62/M	$0.273/M	0%	0.7s	37 tps
GLM 5.1 (Thinking) FP8 z-ai/glm-5.1:thinking:nebius	203k	$1.47/M	$4.62/M	$0.273/M	0%	0.7s	37 tps
GLM 5.2 FP4 z-ai/glm-5.2:nebius	1.0M	$1.47/M	$4.62/M	$0.147/M	N/A	47.8s	3.5 tps
GLM 5.2 (Thinking) FP4 z-ai/glm-5.2:thinking:nebius	1.0M	$1.47/M	$4.62/M	$0.147/M	67%	47.8s	3.5 tps
GLM 5.3 FP4 z-ai/glm-5.3:nebius	1.0M	$1.47/M	$4.62/M	$0.273/M	N/A	0.6s	94 tps
GLM 5.3 (Thinking) FP4 z-ai/glm-5.3:thinking:nebius	1.0M	$1.47/M	$4.62/M	$0.273/M	N/A	0.6s	94 tps
GLM 5.3 Flash FP4 z-ai/glm-5.3-flash:nebius	1.0M	$0.157/M	$0.525/M	$0.0315/M	N/A	0.7s	96 tps
GPT OSS 120B FP4 openai/gpt-oss-120b:nebius	131k	$0.157/M	$0.63/M	$0.105/M	N/A	0.3s	266 tps
Kimi K2.7 Code FP4 moonshotai/kimi-k2.7-code:nebius	262k	$0.998/M	$4.20/M	$0.157/M	N/A	2s	25 tps
Muse Spark Open FP4 meta/muse-spark-open:nebius	1.0M	$1.31/M	$4.46/M	$0.157/M	N/A	0.6s	122 tps

DeepSeek V4 Pro 0813

FP8

deepseek/deepseek-v4-pro-0813:nebius

Context

979k

Input /M

$1.39/M

Output /M

$4.16/M

Cache read /M

$0.042/M

Cache hit

N/A

TTFT

0.8s

Throughput

112 tps

DeepSeek V4 Pro 0813 (Thinking)

FP8

deepseek/deepseek-v4-pro-0813:thinking:nebius

Context

979k

Input /M

$1.39/M

Output /M

$4.16/M

Cache read /M

$0.042/M

Cache hit

N/A

TTFT

0.8s

Throughput

112 tps

DeepSeek V4.1 Flash

FP8

deepseek/deepseek-v4.1-flash:nebius

Context

1.0M

Input /M

$0.315/M

Output /M

$1.26/M

Cache read /M

$0.0105/M

Cache hit

N/A

TTFT

0.7s

Throughput

83 tps

DeepSeek V4.1 Flash (Thinking)

FP8

deepseek/deepseek-v4.1-flash:thinking:nebius

Context

1.0M

Input /M

$0.315/M

Output /M

$1.26/M

Cache read /M

$0.0105/M

Cache hit

N/A

TTFT

0.7s

Throughput

83 tps

DeepSeek V4.1 Flash 1017

FP8

deepseek/deepseek-v4.1-flash-1017:nebius

Context

1.0M

Input /M

$0.315/M

Output /M

$1.26/M

Cache read /M

$0.0063/M

Cache hit

N/A

TTFT

1.6s

Throughput

102 tps

DeepSeek V4.1 Flash 1017 (Thinking)

FP8

deepseek/deepseek-v4.1-flash-1017:thinking:nebius

Context

1.0M

Input /M

$0.315/M

Output /M

$1.26/M

Cache read /M

$0.0063/M

Cache hit

N/A

TTFT

1.6s

Throughput

102 tps

GLM 5.1

FP8

z-ai/glm-5.1:nebius

Context

203k

Input /M

$1.47/M

Output /M

$4.62/M

Cache read /M

$0.273/M

Cache hit

0%

TTFT

0.7s

Throughput

37 tps

GLM 5.1 (Thinking)

FP8

z-ai/glm-5.1:thinking:nebius

Context

203k

Input /M

$1.47/M

Output /M

$4.62/M

Cache read /M

$0.273/M

Cache hit

0%

TTFT

0.7s

Throughput

37 tps

GLM 5.2

FP4

z-ai/glm-5.2:nebius

Context

1.0M

Input /M

$1.47/M

Output /M

$4.62/M

Cache read /M

$0.147/M

Cache hit

N/A

TTFT

47.8s

Throughput

3.5 tps

GLM 5.2 (Thinking)

FP4

z-ai/glm-5.2:thinking:nebius

Context

1.0M

Input /M

$1.47/M

Output /M

$4.62/M

Cache read /M

$0.147/M

Cache hit

67%

TTFT

47.8s

Throughput

3.5 tps

GLM 5.3

FP4

z-ai/glm-5.3:nebius

Context

1.0M

Input /M

$1.47/M

Output /M

$4.62/M

Cache read /M

$0.273/M

Cache hit

N/A

TTFT

0.6s

Throughput

94 tps

GLM 5.3 (Thinking)

FP4

z-ai/glm-5.3:thinking:nebius

Context

1.0M

Input /M

$1.47/M

Output /M

$4.62/M

Cache read /M

$0.273/M

Cache hit

N/A

TTFT

0.6s

Throughput

94 tps

GLM 5.3 Flash

FP4

z-ai/glm-5.3-flash:nebius

Context

1.0M

Input /M

$0.157/M

Output /M

$0.525/M

Cache read /M

$0.0315/M

Cache hit

N/A

TTFT

0.7s

Throughput

96 tps

GPT OSS 120B

FP4

openai/gpt-oss-120b:nebius

Context

131k

Input /M

$0.157/M

Output /M

$0.63/M

Cache read /M

$0.105/M

Cache hit

N/A

TTFT

0.3s

Throughput

266 tps

Kimi K2.7 Code

FP4

moonshotai/kimi-k2.7-code:nebius

Context

262k

Input /M

$0.998/M

Output /M

$4.20/M

Cache read /M

$0.157/M

Cache hit

N/A

TTFT

2s

Throughput

25 tps

Muse Spark Open

FP4

meta/muse-spark-open:nebius

Context

1.0M

Input /M

$1.31/M

Output /M

$4.46/M

Cache read /M

$0.157/M

Cache hit

N/A

TTFT

0.6s

Throughput

122 tps

Prices are per million tokens. TTFT is the observed time to first token. Selectable routes include the applicable provider-selection markup; fixed routes show the standard model price. Availability and pricing refresh continuously; each model page shows the live provider comparison.

Using Nebius via the API

Call a model served by Nebius using its displayed model ID, including the provider suffix when selection is supported.

curl https://nano-gpt.com/api/v1/chat/completions \
  -H "Authorization: Bearer $NANOGPT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek/deepseek-v4-pro-0813:nebius",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

For models with provider selection, append :nebius to the model ID, set "provider": "nebius" in the request body, or send an X-Provider: nebius header. Explicit provider selection adds a route-specific markup over that provider's base price; the prices in the table above already include it. Models without provider selection use the displayed model ID without a provider suffix or selection surcharge.

See the API documentation for provider preferences, price-aware routing, and error behavior.