StreamLake

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 StreamLake

Model name

31 models

Model	Context	Input /M	Output /M	Cache read /M	Cache hit	TTFT	Throughput
DeepSeek R1-0528 deepseek-ai/DeepSeek-R1-0528:streamlake	128k	$0.60/M	$2.40/M	N/A	N/A	3.5s	43 tps
DeepSeek V3.1 Terminus deepseek-ai/DeepSeek-V3.1-Terminus:streamlake	128k	$0.36/M	$1.08/M	N/A	31%	1.1s	52 tps
DeepSeek V3.1 Terminus (Thinking) deepseek-ai/DeepSeek-V3.1-Terminus:thinking:streamlake	128k	$0.36/M	$1.08/M	N/A	N/A	1.1s	52 tps
DeepSeek V4 Flash FP8 deepseek/deepseek-v4-flash:streamlake	1.0M	$0.09/M	$0.17/M	$0.02/M	19.6%	2s	41 tps
DeepSeek V4 Flash (Thinking) FP8 deepseek/deepseek-v4-flash:thinking:streamlake	1.0M	$0.09/M	$0.17/M	$0.02/M	19.6%	2s	41 tps
DeepSeek V4 Flash 0731 FP8 deepseek/deepseek-v4-flash-0731:streamlake	1.0M	$0.14/M	$0.42/M	$0.0044/M	93.2%	1.8s	70 tps
DeepSeek V4 Flash 0731 (Thinking) FP8 deepseek/deepseek-v4-flash-0731:thinking:streamlake	1.0M	$0.14/M	$0.42/M	$0.0044/M	93.2%	1.8s	70 tps
DeepSeek V4 Pro FP8 deepseek/deepseek-v4-pro:streamlake	1.0M	$1.83/M	$3.65/M	$0.15/M	85.3%	1.3s	39 tps
DeepSeek V4 Pro (Thinking) FP8 deepseek/deepseek-v4-pro:thinking:streamlake	1.0M	$1.83/M	$3.65/M	$0.15/M	79.6%	1.3s	39 tps
DeepSeek V4 Pro 0813 deepseek/deepseek-v4-pro-0813:streamlake	1.0M	$0.69/M	$2.08/M	$0.02/M	50.5%	2.2s	51 tps
DeepSeek V4 Pro 0813 (Thinking) deepseek/deepseek-v4-pro-0813:thinking:streamlake	1.0M	$0.69/M	$2.08/M	$0.02/M	48.7%	2.2s	51 tps
DeepSeek V4.1 Flash FP8 deepseek/deepseek-v4.1-flash:streamlake	1.0M	$0.15/M	$0.62/M	$0.0031/M	N/A	1.5s	66 tps
DeepSeek V4.1 Flash (Thinking) FP8 deepseek/deepseek-v4.1-flash:thinking:streamlake	1.0M	$0.15/M	$0.62/M	$0.0031/M	56.3%	1.5s	66 tps
DeepSeek V4.1 Flash 1017 FP8 deepseek/deepseek-v4.1-flash-1017:streamlake	1.0M	$0.16/M	$0.63/M	$0.0032/M	N/A	1.3s	108 tps
DeepSeek V4.1 Flash 1017 (Thinking) FP8 deepseek/deepseek-v4.1-flash-1017:thinking:streamlake	1.0M	$0.16/M	$0.63/M	$0.0032/M	65.8%	1.3s	108 tps
GLM 5 FP8 z-ai/glm-5:streamlake	198k	$0.63/M	$2.02/M	$0.13/M	70.9%	2s	62 tps
GLM 5 (Thinking) FP8 z-ai/glm-5:thinking:streamlake	198k	$0.63/M	$2.02/M	$0.13/M	66.8%	2s	62 tps
GLM 5.1 FP8 z-ai/glm-5.1:streamlake	200k	$1.01/M	$3.19/M	$0.19/M	100%	2.3s	29 tps
GLM 5.1 (Thinking) FP8 z-ai/glm-5.1:thinking:streamlake	200k	$1.01/M	$3.19/M	$0.19/M	100%	2.3s	29 tps
GLM 5.2 FP8 z-ai/glm-5.2:streamlake	1.0M	$0.64/M	$2.02/M	$0.12/M	86.4%	1.7s	45 tps
GLM 5.2 (Thinking) FP8 z-ai/glm-5.2:thinking:streamlake	1.0M	$0.64/M	$2.02/M	$0.12/M	86.7%	1.7s	45 tps
GLM 5.3 Flash FP8 z-ai/glm-5.3-flash:streamlake	1.0M	$0.09/M	$0.31/M	$0.02/M	N/A	1.8s	57 tps
Kimi K2.6 FP8 moonshotai/kimi-k2.6:streamlake	256k	$0.63/M	$2.65/M	$0.11/M	20.3%	0.9s	56 tps
Kimi K2.6 Thinking FP8 moonshotai/kimi-k2.6:thinking:streamlake	256k	$0.63/M	$2.65/M	$0.11/M	20.3%	0.9s	56 tps
Kimi K2.7 Code moonshotai/kimi-k2.7-code:streamlake	256k	$0.75/M	$3.15/M	$0.15/M	19%	1.1s	47 tps
MiMo V2.6 Pro xiaomi/mimo-v2.6-pro:streamlake	1M	$0.55/M	$1.10/M	$0.0045/M	0%	2.6s	23 tps
MiniMax M2.5 minimax/minimax-m2.5:streamlake	200k	$0.28/M	$1.13/M	$0.03/M	N/A	0.6s	113.5 tps
MiniMax M3 FP8 minimax/minimax-m3:streamlake	1M	$0.31/M	$1.26/M	$0.06/M	4.2%	3.9s	69 tps
MiniMax M3 Thinking FP8 minimax/minimax-m3:thinking:streamlake	1M	$0.31/M	$1.26/M	$0.06/M	4.2%	3.9s	69 tps
Qwen3.5 397B qwen/qwen3.5-397b-a17b:streamlake	256k	$0.63/M	$3.78/M	$0.13/M	N/A	1.9s	97 tps
Qwen3.5 397B Thinking qwen/qwen3.5-397b-a17b-thinking:streamlake	256k	$0.63/M	$3.78/M	$0.13/M	N/A	1.9s	97 tps

DeepSeek R1-0528

deepseek-ai/DeepSeek-R1-0528:streamlake

Context

128k

Input /M

$0.60/M

Output /M

$2.40/M

Cache read /M

N/A

Cache hit

N/A

TTFT

3.5s

Throughput

43 tps

DeepSeek V3.1 Terminus

deepseek-ai/DeepSeek-V3.1-Terminus:streamlake

Context

128k

Input /M

$0.36/M

Output /M

$1.08/M

Cache read /M

N/A

Cache hit

31%

TTFT

1.1s

Throughput

52 tps

DeepSeek V3.1 Terminus (Thinking)

deepseek-ai/DeepSeek-V3.1-Terminus:thinking:streamlake

Context

128k

Input /M

$0.36/M

Output /M

$1.08/M

Cache read /M

N/A

Cache hit

N/A

TTFT

1.1s

Throughput

52 tps

DeepSeek V4 Flash

FP8

deepseek/deepseek-v4-flash:streamlake

Context

1.0M

Input /M

$0.09/M

Output /M

$0.17/M

Cache read /M

$0.02/M

Cache hit

19.6%

TTFT

2s

Throughput

41 tps

DeepSeek V4 Flash (Thinking)

FP8

deepseek/deepseek-v4-flash:thinking:streamlake

Context

1.0M

Input /M

$0.09/M

Output /M

$0.17/M

Cache read /M

$0.02/M

Cache hit

19.6%

TTFT

2s

Throughput

41 tps

DeepSeek V4 Flash 0731

FP8

deepseek/deepseek-v4-flash-0731:streamlake

Context

1.0M

Input /M

$0.14/M

Output /M

$0.42/M

Cache read /M

$0.0044/M

Cache hit

93.2%

TTFT

1.8s

Throughput

70 tps

DeepSeek V4 Flash 0731 (Thinking)

FP8

deepseek/deepseek-v4-flash-0731:thinking:streamlake

Context

1.0M

Input /M

$0.14/M

Output /M

$0.42/M

Cache read /M

$0.0044/M

Cache hit

93.2%

TTFT

1.8s

Throughput

70 tps

DeepSeek V4 Pro

FP8

deepseek/deepseek-v4-pro:streamlake

Context

1.0M

Input /M

$1.83/M

Output /M

$3.65/M

Cache read /M

$0.15/M

Cache hit

85.3%

TTFT

1.3s

Throughput

39 tps

DeepSeek V4 Pro (Thinking)

FP8

deepseek/deepseek-v4-pro:thinking:streamlake

Context

1.0M

Input /M

$1.83/M

Output /M

$3.65/M

Cache read /M

$0.15/M

Cache hit

79.6%

TTFT

1.3s

Throughput

39 tps

DeepSeek V4 Pro 0813

deepseek/deepseek-v4-pro-0813:streamlake

Context

1.0M

Input /M

$0.69/M

Output /M

$2.08/M

Cache read /M

$0.02/M

Cache hit

50.5%

TTFT

2.2s

Throughput

51 tps

DeepSeek V4 Pro 0813 (Thinking)

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

Context

1.0M

Input /M

$0.69/M

Output /M

$2.08/M

Cache read /M

$0.02/M

Cache hit

48.7%

TTFT

2.2s

Throughput

51 tps

DeepSeek V4.1 Flash

FP8

deepseek/deepseek-v4.1-flash:streamlake

Context

1.0M

Input /M

$0.15/M

Output /M

$0.62/M

Cache read /M

$0.0031/M

Cache hit

N/A

TTFT

1.5s

Throughput

66 tps

DeepSeek V4.1 Flash (Thinking)

FP8

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

Context

1.0M

Input /M

$0.15/M

Output /M

$0.62/M

Cache read /M

$0.0031/M

Cache hit

56.3%

TTFT

1.5s

Throughput

66 tps

DeepSeek V4.1 Flash 1017

FP8

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

Context

1.0M

Input /M

$0.16/M

Output /M

$0.63/M

Cache read /M

$0.0032/M

Cache hit

N/A

TTFT

1.3s

Throughput

108 tps

DeepSeek V4.1 Flash 1017 (Thinking)

FP8

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

Context

1.0M

Input /M

$0.16/M

Output /M

$0.63/M

Cache read /M

$0.0032/M

Cache hit

65.8%

TTFT

1.3s

Throughput

108 tps

GLM 5

FP8

z-ai/glm-5:streamlake

Context

198k

Input /M

$0.63/M

Output /M

$2.02/M

Cache read /M

$0.13/M

Cache hit

70.9%

TTFT

2s

Throughput

62 tps

GLM 5 (Thinking)

FP8

z-ai/glm-5:thinking:streamlake

Context

198k

Input /M

$0.63/M

Output /M

$2.02/M

Cache read /M

$0.13/M

Cache hit

66.8%

TTFT

2s

Throughput

62 tps

GLM 5.1

FP8

z-ai/glm-5.1:streamlake

Context

200k

Input /M

$1.01/M

Output /M

$3.19/M

Cache read /M

$0.19/M

Cache hit

100%

TTFT

2.3s

Throughput

29 tps

GLM 5.1 (Thinking)

FP8

z-ai/glm-5.1:thinking:streamlake

Context

200k

Input /M

$1.01/M

Output /M

$3.19/M

Cache read /M

$0.19/M

Cache hit

100%

TTFT

2.3s

Throughput

29 tps

GLM 5.2

FP8

z-ai/glm-5.2:streamlake

Context

1.0M

Input /M

$0.64/M

Output /M

$2.02/M

Cache read /M

$0.12/M

Cache hit

86.4%

TTFT

1.7s

Throughput

45 tps

GLM 5.2 (Thinking)

FP8

z-ai/glm-5.2:thinking:streamlake

Context

1.0M

Input /M

$0.64/M

Output /M

$2.02/M

Cache read /M

$0.12/M

Cache hit

86.7%

TTFT

1.7s

Throughput

45 tps

GLM 5.3 Flash

FP8

z-ai/glm-5.3-flash:streamlake

Context

1.0M

Input /M

$0.09/M

Output /M

$0.31/M

Cache read /M

$0.02/M

Cache hit

N/A

TTFT

1.8s

Throughput

57 tps

Kimi K2.6

FP8

moonshotai/kimi-k2.6:streamlake

Context

256k

Input /M

$0.63/M

Output /M

$2.65/M

Cache read /M

$0.11/M

Cache hit

20.3%

TTFT

0.9s

Throughput

56 tps

Kimi K2.6 Thinking

FP8

moonshotai/kimi-k2.6:thinking:streamlake

Context

256k

Input /M

$0.63/M

Output /M

$2.65/M

Cache read /M

$0.11/M

Cache hit

20.3%

TTFT

0.9s

Throughput

56 tps

Kimi K2.7 Code

moonshotai/kimi-k2.7-code:streamlake

Context

256k

Input /M

$0.75/M

Output /M

$3.15/M

Cache read /M

$0.15/M

Cache hit

19%

TTFT

1.1s

Throughput

47 tps

MiMo V2.6 Pro

xiaomi/mimo-v2.6-pro:streamlake

Context

1M

Input /M

$0.55/M

Output /M

$1.10/M

Cache read /M

$0.0045/M

Cache hit

0%

TTFT

2.6s

Throughput

23 tps

MiniMax M2.5

minimax/minimax-m2.5:streamlake

Context

200k

Input /M

$0.28/M

Output /M

$1.13/M

Cache read /M

$0.03/M

Cache hit

N/A

TTFT

0.6s

Throughput

113.5 tps

MiniMax M3

FP8

minimax/minimax-m3:streamlake

Context

1M

Input /M

$0.31/M

Output /M

$1.26/M

Cache read /M

$0.06/M

Cache hit

4.2%

TTFT

3.9s

Throughput

69 tps

MiniMax M3 Thinking

FP8

minimax/minimax-m3:thinking:streamlake

Context

1M

Input /M

$0.31/M

Output /M

$1.26/M

Cache read /M

$0.06/M

Cache hit

4.2%

TTFT

3.9s

Throughput

69 tps

Qwen3.5 397B

qwen/qwen3.5-397b-a17b:streamlake

Context

256k

Input /M

$0.63/M

Output /M

$3.78/M

Cache read /M

$0.13/M

Cache hit

N/A

TTFT

1.9s

Throughput

97 tps

Qwen3.5 397B Thinking

qwen/qwen3.5-397b-a17b-thinking:streamlake

Context

256k

Input /M

$0.63/M

Output /M

$3.78/M

Cache read /M

$0.13/M

Cache hit

N/A

TTFT

1.9s

Throughput

97 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 StreamLake via the API

Call a model served by StreamLake 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-ai/DeepSeek-R1-0528:streamlake",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

For models with provider selection, append :streamlake to the model ID, set "provider": "streamlake" in the request body, or send an X-Provider: streamlake 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.