Phala

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 Phala

Model name

26 models

Model	Context	Input /M	Output /M	Cache read /M	TTFT	Throughput
DeepSeek V3.2 deepseek/deepseek-v3.2:phala	164k	$1.05/M	$1.05/M	$0.53/M	3.2s	13.5 tps
DeepSeek V3.2 (Thinking) deepseek/deepseek-v3.2:thinking:phala	164k	$1.05/M	$1.05/M	$0.53/M	3.2s	13.5 tps
DeepSeek V4 Flash 0731 deepseek/deepseek-v4-flash-0731:phala	1.0M	$0.32/M	$0.97/M	$0.02/M	2.4s	68 tps
DeepSeek V4 Flash 0731 (Thinking) deepseek/deepseek-v4-flash-0731:thinking:phala	1.0M	$0.32/M	$0.97/M	$0.02/M	2.4s	68 tps
DeepSeek V4 Pro 0813 deepseek/deepseek-v4-pro-0813:phala	1.0M	$1.00/M	$3.02/M	$0.10/M	3.2s	47.5 tps
DeepSeek V4 Pro 0813 (Thinking) deepseek/deepseek-v4-pro-0813:thinking:phala	1.0M	$1.00/M	$3.02/M	$0.10/M	3.2s	47.5 tps
DeepSeek V4.1 Flash deepseek/deepseek-v4.1-flash:phala	1.0M	$0.22/M	$0.88/M	$0.0044/M	1.3s	55 tps
DeepSeek V4.1 Flash (Thinking) deepseek/deepseek-v4.1-flash:thinking:phala	1.0M	$0.22/M	$0.88/M	$0.0044/M	1.3s	55 tps
GLM 5.2 FP8 z-ai/glm-5.2:phala	1.0M	$1.32/M	$3.15/M	$0.23/M	2.4s	68.5 tps
GLM 5.2 (Thinking) FP8 z-ai/glm-5.2:thinking:phala	1.0M	$1.32/M	$3.15/M	$0.23/M	2.4s	68.5 tps
GLM 5.3 z-ai/glm-5.3:phala	1.0M	$0.88/M	$2.77/M	$0.16/M	3.7s	49 tps
GLM 5.3 (Thinking) z-ai/glm-5.3:thinking:phala	1.0M	$0.88/M	$2.77/M	$0.16/M	3.7s	49 tps
GLM 5.3 Flash NVFP4 z-ai/glm-5.3-flash:phala	1.0M	$0.12/M	$0.39/M	$0.02/M	3.8s	37 tps
GPT OSS 120B openai/gpt-oss-120b:phala	131k	$0.16/M	$0.63/M	N/A	0.8s	147 tps
Kimi K2.7 Code moonshotai/kimi-k2.7-code:phala	262k	$1.00/M	$4.20/M	$0.20/M	1.6s	66 tps
Kimi K3 moonshotai/kimi-k3:phala	1.0M	$2.68/M	$13.39/M	$0.27/M	3.3s	40 tps
Qwen 3.6 27B alibaba/qwen3.6-27b:phala	262k	$0.34/M	$3.41/M	$0.03/M	5.5s	9 tps
Qwen 3.6 27B Thinking alibaba/qwen3.6-27b:thinking:phala	262k	$0.34/M	$3.41/M	$0.03/M	5.5s	9 tps
Qwen 3.6 35B A3B Qwen/Qwen3.6-35B-A3B:phala	262k	$0.21/M	$1.33/M	$0.06/M	1.4s	55 tps
Qwen 3.6 35B A3B Thinking Qwen/Qwen3.6-35B-A3B:thinking:phala	262k	$0.21/M	$1.33/M	$0.06/M	1.4s	55 tps
Qwen3.5 27B qwen3.5-27b:phala	262k	$0.31/M	$2.52/M	$0.03/M	2.3s	19 tps
Qwen3.5 27B Thinking qwen3.5-27b:thinking:phala	262k	$0.31/M	$2.52/M	$0.03/M	2.3s	19 tps
Qwen3.5 397B qwen/qwen3.5-397b-a17b:phala	262k	$0.58/M	$3.68/M	$0.24/M	1.3s	46 tps
Qwen3.5 397B Thinking qwen/qwen3.5-397b-a17b-thinking:phala	262k	$0.58/M	$3.68/M	$0.24/M	1.3s	46 tps
Qwen3.8 27B qwen3.8-27b:phala	1M	$0.21/M	$2.63/M	$0.05/M	1.7s	46 tps
Qwen3.8 27B Thinking qwen3.8-27b:thinking:phala	1M	$0.21/M	$2.63/M	$0.05/M	1.7s	46 tps

DeepSeek V3.2

deepseek/deepseek-v3.2:phala

Context

164k

Input /M

$1.05/M

Output /M

$1.05/M

Cache read /M

$0.53/M

TTFT

3.2s

Throughput

13.5 tps

DeepSeek V3.2 (Thinking)

deepseek/deepseek-v3.2:thinking:phala

Context

164k

Input /M

$1.05/M

Output /M

$1.05/M

Cache read /M

$0.53/M

TTFT

3.2s

Throughput

13.5 tps

DeepSeek V4 Flash 0731

deepseek/deepseek-v4-flash-0731:phala

Context

1.0M

Input /M

$0.32/M

Output /M

$0.97/M

Cache read /M

$0.02/M

TTFT

2.4s

Throughput

68 tps

DeepSeek V4 Flash 0731 (Thinking)

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

Context

1.0M

Input /M

$0.32/M

Output /M

$0.97/M

Cache read /M

$0.02/M

TTFT

2.4s

Throughput

68 tps

DeepSeek V4 Pro 0813

deepseek/deepseek-v4-pro-0813:phala

Context

1.0M

Input /M

$1.00/M

Output /M

$3.02/M

Cache read /M

$0.10/M

TTFT

3.2s

Throughput

47.5 tps

DeepSeek V4 Pro 0813 (Thinking)

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

Context

1.0M

Input /M

$1.00/M

Output /M

$3.02/M

Cache read /M

$0.10/M

TTFT

3.2s

Throughput

47.5 tps

DeepSeek V4.1 Flash

deepseek/deepseek-v4.1-flash:phala

Context

1.0M

Input /M

$0.22/M

Output /M

$0.88/M

Cache read /M

$0.0044/M

TTFT

1.3s

Throughput

55 tps

DeepSeek V4.1 Flash (Thinking)

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

Context

1.0M

Input /M

$0.22/M

Output /M

$0.88/M

Cache read /M

$0.0044/M

TTFT

1.3s

Throughput

55 tps

GLM 5.2

FP8

z-ai/glm-5.2:phala

Context

1.0M

Input /M

$1.32/M

Output /M

$3.15/M

Cache read /M

$0.23/M

TTFT

2.4s

Throughput

68.5 tps

GLM 5.2 (Thinking)

FP8

z-ai/glm-5.2:thinking:phala

Context

1.0M

Input /M

$1.32/M

Output /M

$3.15/M

Cache read /M

$0.23/M

TTFT

2.4s

Throughput

68.5 tps

GLM 5.3

z-ai/glm-5.3:phala

Context

1.0M

Input /M

$0.88/M

Output /M

$2.77/M

Cache read /M

$0.16/M

TTFT

3.7s

Throughput

49 tps

GLM 5.3 (Thinking)

z-ai/glm-5.3:thinking:phala

Context

1.0M

Input /M

$0.88/M

Output /M

$2.77/M

Cache read /M

$0.16/M

TTFT

3.7s

Throughput

49 tps

GLM 5.3 Flash

NVFP4

z-ai/glm-5.3-flash:phala

Context

1.0M

Input /M

$0.12/M

Output /M

$0.39/M

Cache read /M

$0.02/M

TTFT

3.8s

Throughput

37 tps

GPT OSS 120B

openai/gpt-oss-120b:phala

Context

131k

Input /M

$0.16/M

Output /M

$0.63/M

Cache read /M

N/A

TTFT

0.8s

Throughput

147 tps

Kimi K2.7 Code

moonshotai/kimi-k2.7-code:phala

Context

262k

Input /M

$1.00/M

Output /M

$4.20/M

Cache read /M

$0.20/M

TTFT

1.6s

Throughput

66 tps

Kimi K3

moonshotai/kimi-k3:phala

Context

1.0M

Input /M

$2.68/M

Output /M

$13.39/M

Cache read /M

$0.27/M

TTFT

3.3s

Throughput

40 tps

Qwen 3.6 27B

alibaba/qwen3.6-27b:phala

Context

262k

Input /M

$0.34/M

Output /M

$3.41/M

Cache read /M

$0.03/M

TTFT

5.5s

Throughput

9 tps

Qwen 3.6 27B Thinking

alibaba/qwen3.6-27b:thinking:phala

Context

262k

Input /M

$0.34/M

Output /M

$3.41/M

Cache read /M

$0.03/M

TTFT

5.5s

Throughput

9 tps

Qwen 3.6 35B A3B

Qwen/Qwen3.6-35B-A3B:phala

Context

262k

Input /M

$0.21/M

Output /M

$1.33/M

Cache read /M

$0.06/M

TTFT

1.4s

Throughput

55 tps

Qwen 3.6 35B A3B Thinking

Qwen/Qwen3.6-35B-A3B:thinking:phala

Context

262k

Input /M

$0.21/M

Output /M

$1.33/M

Cache read /M

$0.06/M

TTFT

1.4s

Throughput

55 tps

Qwen3.5 27B

qwen3.5-27b:phala

Context

262k

Input /M

$0.31/M

Output /M

$2.52/M

Cache read /M

$0.03/M

TTFT

2.3s

Throughput

19 tps

Qwen3.5 27B Thinking

qwen3.5-27b:thinking:phala

Context

262k

Input /M

$0.31/M

Output /M

$2.52/M

Cache read /M

$0.03/M

TTFT

2.3s

Throughput

19 tps

Qwen3.5 397B

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

Context

262k

Input /M

$0.58/M

Output /M

$3.68/M

Cache read /M

$0.24/M

TTFT

1.3s

Throughput

46 tps

Qwen3.5 397B Thinking

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

Context

262k

Input /M

$0.58/M

Output /M

$3.68/M

Cache read /M

$0.24/M

TTFT

1.3s

Throughput

46 tps

Qwen3.8 27B

qwen3.8-27b:phala

Context

1M

Input /M

$0.21/M

Output /M

$2.63/M

Cache read /M

$0.05/M

TTFT

1.7s

Throughput

46 tps

Qwen3.8 27B Thinking

qwen3.8-27b:thinking:phala

Context

1M

Input /M

$0.21/M

Output /M

$2.63/M

Cache read /M

$0.05/M

TTFT

1.7s

Throughput

46 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 Phala via the API

Call a model served by Phala 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-v3.2:phala",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

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