Parasail

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 Parasail

Model name

43 models

Model	Context	Input /M	Output /M	Cache read /M	TTFT	Throughput
DeepSeek V4 Flash FP8 deepseek/deepseek-v4-flash:parasail	1.0M	$0.15/M	$0.29/M	$0.07/M	0.7s	82 tps
DeepSeek V4 Flash (Thinking) FP8 deepseek/deepseek-v4-flash:thinking:parasail	1.0M	$0.15/M	$0.29/M	$0.07/M	0.7s	82 tps
DeepSeek V4 Flash 0731 FP8 deepseek/deepseek-v4-flash-0731:parasail	1.0M	$0.15/M	$0.29/M	$0.05/M	0.6s	75 tps
DeepSeek V4 Flash 0731 (Thinking) FP8 deepseek/deepseek-v4-flash-0731:thinking:parasail	1.0M	$0.15/M	$0.29/M	$0.05/M	0.6s	75 tps
DeepSeek V4 Pro FP8 deepseek/deepseek-v4-pro:parasail	1.0M	$1.83/M	$3.65/M	$0.11/M	1.2s	34 tps
DeepSeek V4 Pro (Thinking) FP8 deepseek/deepseek-v4-pro:thinking:parasail	1.0M	$1.83/M	$3.65/M	$0.11/M	1.2s	34 tps
DeepSeek V4 Pro 0813 FP8 deepseek/deepseek-v4-pro-0813:parasail	1.0M	$1.39/M	$4.16/M	$0.05/M	1s	67 tps
DeepSeek V4 Pro 0813 (Thinking) FP8 deepseek/deepseek-v4-pro-0813:thinking:parasail	1.0M	$1.39/M	$4.16/M	$0.05/M	1s	67 tps
DeepSeek V4.1 Flash FP8 deepseek/deepseek-v4.1-flash:parasail	1.0M	$0.31/M	$1.26/M	$0.0063/M	0.7s	108 tps
DeepSeek V4.1 Flash (Thinking) FP8 deepseek/deepseek-v4.1-flash:thinking:parasail	1.0M	$0.31/M	$1.26/M	$0.0063/M	0.7s	108 tps
Gemma 4 26B A4B BF16 google/gemma-4-26b-a4b-it:parasail	262k	$0.14/M	$0.42/M	$0.05/M	0.6s	58 tps
Gemma 4 26B A4B (Thinking) BF16 google/gemma-4-26b-a4b-it:thinking:parasail	262k	$0.14/M	$0.42/M	$0.05/M	0.6s	58 tps
Gemma 4 31B FP8 google/gemma-4-31b-it:parasail	262k	$0.16/M	$0.42/M	$0.06/M	1.5s	20 tps
Gemma 4 31B (Thinking) FP8 google/gemma-4-31b-it:thinking:parasail	262k	$0.16/M	$0.42/M	$0.06/M	1.5s	20 tps
GLM 5.2 (Thinking) FP4 z-ai/glm-5.2:thinking:parasail	1.0M	$1.16/M	$3.47/M	$0.21/M	0.6s	97 tps
GLM 5.3 FP8 z-ai/glm-5.3:parasail	1.0M	$1.47/M	$4.62/M	$0.27/M	2.6s	51 tps
GLM 5.3 (Thinking) FP8 z-ai/glm-5.3:thinking:parasail	1.0M	$1.47/M	$4.62/M	$0.27/M	2.6s	51 tps
GLM 5.3 Flash FP4 z-ai/glm-5.3-flash:parasail	1.0M	$0.16/M	$0.53/M	$0.03/M	0.7s	126 tps
GPT OSS 120B FP4 openai/gpt-oss-120b:parasail	131k	$0.10/M	$0.79/M	$0.06/M	0.4s	104 tps
GPT OSS 20B FP4 openai/gpt-oss-20b:parasail	131k	$0.03/M	$0.16/M	$0.02/M	1.6s	33 tps
Kimi K2.6 INT4 moonshotai/kimi-k2.6:parasail	262k	$0.79/M	$3.68/M	$0.17/M	2.1s	46 tps
Kimi K2.6 Thinking INT4 moonshotai/kimi-k2.6:thinking:parasail	262k	$0.79/M	$3.68/M	$0.17/M	2.1s	46 tps
Kimi K3 FP4 moonshotai/kimi-k3:parasail	1.0M	$2.73/M	$13.65/M	$0.27/M	1.3s	73 tps
Llama 3.3 70B Instruct FP8 meta-llama/llama-3.3-70b-instruct:parasail	131k	$0.23/M	$0.53/M	$0.12/M	0.6s	52 tps
Llama 4 Maverick FP8 meta-llama/llama-4-maverick:parasail	524k	$0.37/M	$1.05/M	$0.18/M	0.4s	81 tps
MiniMax M3 FP8 minimax/minimax-m3:parasail	1.0M	$0.31/M	$1.26/M	$0.06/M	0.9s	97 tps
MiniMax M3 Thinking FP8 minimax/minimax-m3:thinking:parasail	1.0M	$0.31/M	$1.26/M	$0.06/M	0.9s	97 tps
Phi 4 Multimodal phi-4-multimodal-instruct	128k	$0.07/M	$0.11/M	N/A	N/A	N/A
Qwen 3.6 35B A3B FP8 Qwen/Qwen3.6-35B-A3B:parasail	262k	$0.16/M	$1.05/M	$0.05/M	1.1s	35 tps
Qwen 3.6 35B A3B Thinking FP8 Qwen/Qwen3.6-35B-A3B:thinking:parasail	262k	$0.16/M	$1.05/M	$0.05/M	1.1s	35 tps
Qwen25 VL 72b qwen25-vl-72b-instruct	32k	$0.70/M	$0.70/M	N/A	N/A	N/A
Qwen3 235B A22B FP8 qwen/qwen3-235b-a22b:parasail	131k	$0.15/M	$0.84/M	$0.05/M	0.5s	37 tps
Qwen3 Coder Next BF16 qwen/qwen3-coder-next:parasail	262k	$0.13/M	$0.84/M	$0.07/M	0.7s	49 tps
Qwen3 Next 80B A3B Instruct FP8 Qwen/Qwen3-Next-80B-A3B-Instruct:parasail	262k	$0.10/M	$1.16/M	$0.07/M	0.6s	124 tps
Qwen3.5 35B A3B FP8 qwen3.5-35b-a3b:parasail	260k	$0.24/M	$1.89/M	Supported	0.4s	122 tps
Qwen3.5 397B FP8 qwen/qwen3.5-397b-a17b:parasail	262k	$0.53/M	$3.78/M	$0.31/M	0.8s	37 tps
Qwen3.5 397B Thinking FP8 qwen/qwen3.5-397b-a17b-thinking:parasail	262k	$0.53/M	$3.78/M	$0.31/M	0.8s	37 tps
Qwen3.5 9B BF16 qwen/qwen3.5-9b:parasail	262k	$0.10/M	$0.26/M	N/A	0.5s	97 tps
Qwen3.8 27B FP8 qwen3.8-27b:parasail	262k	$0.25/M	$2.31/M	$0.05/M	0.7s	47 tps
Qwen3.8 27B Thinking FP8 qwen3.8-27b:thinking:parasail	262k	$0.25/M	$2.31/M	$0.05/M	0.7s	47 tps
Steelskull Electra R1 70b Steelskull/L3.3-Electra-R1-70b	16k	$0.70/M	$0.70/M	N/A	N/A	N/A
The Drummer Cydonia 24B v4.1 BF16 TheDrummer/Cydonia-24B-v4.1:parasail	131k	$0.31/M	$0.53/M	$0.16/M	0.8s	42 tps
TheDrummer Skyfall 36B V2 FP8 thedrummer/skyfall-36b-v2:parasail	33k	$0.58/M	$0.84/M	$0.26/M	0.5s	67 tps

DeepSeek V4 Flash

FP8

deepseek/deepseek-v4-flash:parasail

Context

1.0M

Input /M

$0.15/M

Output /M

$0.29/M

Cache read /M

$0.07/M

TTFT

0.7s

Throughput

82 tps

DeepSeek V4 Flash (Thinking)

FP8

deepseek/deepseek-v4-flash:thinking:parasail

Context

1.0M

Input /M

$0.15/M

Output /M

$0.29/M

Cache read /M

$0.07/M

TTFT

0.7s

Throughput

82 tps

DeepSeek V4 Flash 0731

FP8

deepseek/deepseek-v4-flash-0731:parasail

Context

1.0M

Input /M

$0.15/M

Output /M

$0.29/M

Cache read /M

$0.05/M

TTFT

0.6s

Throughput

75 tps

DeepSeek V4 Flash 0731 (Thinking)

FP8

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

Context

1.0M

Input /M

$0.15/M

Output /M

$0.29/M

Cache read /M

$0.05/M

TTFT

0.6s

Throughput

75 tps

DeepSeek V4 Pro

FP8

deepseek/deepseek-v4-pro:parasail

Context

1.0M

Input /M

$1.83/M

Output /M

$3.65/M

Cache read /M

$0.11/M

TTFT

1.2s

Throughput

34 tps

DeepSeek V4 Pro (Thinking)

FP8

deepseek/deepseek-v4-pro:thinking:parasail

Context

1.0M

Input /M

$1.83/M

Output /M

$3.65/M

Cache read /M

$0.11/M

TTFT

1.2s

Throughput

34 tps

DeepSeek V4 Pro 0813

FP8

deepseek/deepseek-v4-pro-0813:parasail

Context

1.0M

Input /M

$1.39/M

Output /M

$4.16/M

Cache read /M

$0.05/M

TTFT

1s

Throughput

67 tps

DeepSeek V4 Pro 0813 (Thinking)

FP8

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

Context

1.0M

Input /M

$1.39/M

Output /M

$4.16/M

Cache read /M

$0.05/M

TTFT

1s

Throughput

67 tps

DeepSeek V4.1 Flash

FP8

deepseek/deepseek-v4.1-flash:parasail

Context

1.0M

Input /M

$0.31/M

Output /M

$1.26/M

Cache read /M

$0.0063/M

TTFT

0.7s

Throughput

108 tps

DeepSeek V4.1 Flash (Thinking)

FP8

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

Context

1.0M

Input /M

$0.31/M

Output /M

$1.26/M

Cache read /M

$0.0063/M

TTFT

0.7s

Throughput

108 tps

Gemma 4 26B A4B

BF16

google/gemma-4-26b-a4b-it:parasail

Context

262k

Input /M

$0.14/M

Output /M

$0.42/M

Cache read /M

$0.05/M

TTFT

0.6s

Throughput

58 tps

Gemma 4 26B A4B (Thinking)

BF16

google/gemma-4-26b-a4b-it:thinking:parasail

Context

262k

Input /M

$0.14/M

Output /M

$0.42/M

Cache read /M

$0.05/M

TTFT

0.6s

Throughput

58 tps

Gemma 4 31B

FP8

google/gemma-4-31b-it:parasail

Context

262k

Input /M

$0.16/M

Output /M

$0.42/M

Cache read /M

$0.06/M

TTFT

1.5s

Throughput

20 tps

Gemma 4 31B (Thinking)

FP8

google/gemma-4-31b-it:thinking:parasail

Context

262k

Input /M

$0.16/M

Output /M

$0.42/M

Cache read /M

$0.06/M

TTFT

1.5s

Throughput

20 tps

GLM 5.2 (Thinking)

FP4

z-ai/glm-5.2:thinking:parasail

Context

1.0M

Input /M

$1.16/M

Output /M

$3.47/M

Cache read /M

$0.21/M

TTFT

0.6s

Throughput

97 tps

GLM 5.3

FP8

z-ai/glm-5.3:parasail

Context

1.0M

Input /M

$1.47/M

Output /M

$4.62/M

Cache read /M

$0.27/M

TTFT

2.6s

Throughput

51 tps

GLM 5.3 (Thinking)

FP8

z-ai/glm-5.3:thinking:parasail

Context

1.0M

Input /M

$1.47/M

Output /M

$4.62/M

Cache read /M

$0.27/M

TTFT

2.6s

Throughput

51 tps

GLM 5.3 Flash

FP4

z-ai/glm-5.3-flash:parasail

Context

1.0M

Input /M

$0.16/M

Output /M

$0.53/M

Cache read /M

$0.03/M

TTFT

0.7s

Throughput

126 tps

GPT OSS 120B

FP4

openai/gpt-oss-120b:parasail

Context

131k

Input /M

$0.10/M

Output /M

$0.79/M

Cache read /M

$0.06/M

TTFT

0.4s

Throughput

104 tps

GPT OSS 20B

FP4

openai/gpt-oss-20b:parasail

Context

131k

Input /M

$0.03/M

Output /M

$0.16/M

Cache read /M

$0.02/M

TTFT

1.6s

Throughput

33 tps

Kimi K2.6

INT4

moonshotai/kimi-k2.6:parasail

Context

262k

Input /M

$0.79/M

Output /M

$3.68/M

Cache read /M

$0.17/M

TTFT

2.1s

Throughput

46 tps

Kimi K2.6 Thinking

INT4

moonshotai/kimi-k2.6:thinking:parasail

Context

262k

Input /M

$0.79/M

Output /M

$3.68/M

Cache read /M

$0.17/M

TTFT

2.1s

Throughput

46 tps

Kimi K3

FP4

moonshotai/kimi-k3:parasail

Context

1.0M

Input /M

$2.73/M

Output /M

$13.65/M

Cache read /M

$0.27/M

TTFT

1.3s

Throughput

73 tps

Llama 3.3 70B Instruct

FP8

meta-llama/llama-3.3-70b-instruct:parasail

Context

131k

Input /M

$0.23/M

Output /M

$0.53/M

Cache read /M

$0.12/M

TTFT

0.6s

Throughput

52 tps

Llama 4 Maverick

FP8

meta-llama/llama-4-maverick:parasail

Context

524k

Input /M

$0.37/M

Output /M

$1.05/M

Cache read /M

$0.18/M

TTFT

0.4s

Throughput

81 tps

MiniMax M3

FP8

minimax/minimax-m3:parasail

Context

1.0M

Input /M

$0.31/M

Output /M

$1.26/M

Cache read /M

$0.06/M

TTFT

0.9s

Throughput

97 tps

MiniMax M3 Thinking

FP8

minimax/minimax-m3:thinking:parasail

Context

1.0M

Input /M

$0.31/M

Output /M

$1.26/M

Cache read /M

$0.06/M

TTFT

0.9s

Throughput

97 tps

Phi 4 Multimodal

phi-4-multimodal-instruct

Context

128k

Input /M

$0.07/M

Output /M

$0.11/M

Cache read /M

N/A

TTFT

N/A

Throughput

N/A

Qwen 3.6 35B A3B

FP8

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

Context

262k

Input /M

$0.16/M

Output /M

$1.05/M

Cache read /M

$0.05/M

TTFT

1.1s

Throughput

35 tps

Qwen 3.6 35B A3B Thinking

FP8

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

Context

262k

Input /M

$0.16/M

Output /M

$1.05/M

Cache read /M

$0.05/M

TTFT

1.1s

Throughput

35 tps

Qwen25 VL 72b

qwen25-vl-72b-instruct

Context

32k

Input /M

$0.70/M

Output /M

$0.70/M

Cache read /M

N/A

TTFT

N/A

Throughput

N/A

Qwen3 235B A22B

FP8

qwen/qwen3-235b-a22b:parasail

Context

131k

Input /M

$0.15/M

Output /M

$0.84/M

Cache read /M

$0.05/M

TTFT

0.5s

Throughput

37 tps

Qwen3 Coder Next

BF16

qwen/qwen3-coder-next:parasail

Context

262k

Input /M

$0.13/M

Output /M

$0.84/M

Cache read /M

$0.07/M

TTFT

0.7s

Throughput

49 tps

Qwen3 Next 80B A3B Instruct

FP8

Qwen/Qwen3-Next-80B-A3B-Instruct:parasail

Context

262k

Input /M

$0.10/M

Output /M

$1.16/M

Cache read /M

$0.07/M

TTFT

0.6s

Throughput

124 tps

Qwen3.5 35B A3B

FP8

qwen3.5-35b-a3b:parasail

Context

260k

Input /M

$0.24/M

Output /M

$1.89/M

Cache read /M

Supported

TTFT

0.4s

Throughput

122 tps

Qwen3.5 397B

FP8

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

Context

262k

Input /M

$0.53/M

Output /M

$3.78/M

Cache read /M

$0.31/M

TTFT

0.8s

Throughput

37 tps

Qwen3.5 397B Thinking

FP8

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

Context

262k

Input /M

$0.53/M

Output /M

$3.78/M

Cache read /M

$0.31/M

TTFT

0.8s

Throughput

37 tps

Qwen3.5 9B

BF16

qwen/qwen3.5-9b:parasail

Context

262k

Input /M

$0.10/M

Output /M

$0.26/M

Cache read /M

N/A

TTFT

0.5s

Throughput

97 tps

Qwen3.8 27B

FP8

qwen3.8-27b:parasail

Context

262k

Input /M

$0.25/M

Output /M

$2.31/M

Cache read /M

$0.05/M

TTFT

0.7s

Throughput

47 tps

Qwen3.8 27B Thinking

FP8

qwen3.8-27b:thinking:parasail

Context

262k

Input /M

$0.25/M

Output /M

$2.31/M

Cache read /M

$0.05/M

TTFT

0.7s

Throughput

47 tps

Steelskull Electra R1 70b

Steelskull/L3.3-Electra-R1-70b

Context

16k

Input /M

$0.70/M

Output /M

$0.70/M

Cache read /M

N/A

TTFT

N/A

Throughput

N/A

The Drummer Cydonia 24B v4.1

BF16

TheDrummer/Cydonia-24B-v4.1:parasail

Context

131k

Input /M

$0.31/M

Output /M

$0.53/M

Cache read /M

$0.16/M

TTFT

0.8s

Throughput

42 tps

TheDrummer Skyfall 36B V2

FP8

thedrummer/skyfall-36b-v2:parasail

Context

33k

Input /M

$0.58/M

Output /M

$0.84/M

Cache read /M

$0.26/M

TTFT

0.5s

Throughput

67 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 Parasail via the API

Call a model served by Parasail 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-flash:parasail",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

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