Friendli

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 Friendli

Model name

10 models

Model	Context	Input /M	Output /M	Cache read /M	TTFT	Throughput
DeepSeek V3.2 deepseek/deepseek-v3.2:friendli	164k	$0.53/M	$1.57/M	$0.26/M	2.4s	35 tps
Gemma 4 31B google/gemma-4-31b-it:friendli	262k	$0.15/M	$0.42/M	N/A	0.6s	86 tps
Gemma 4 31B (Thinking) google/gemma-4-31b-it:thinking:friendli	262k	$0.15/M	$0.42/M	N/A	0.6s	86 tps
GLM 5 z-ai/glm-5:friendli	200k	$1.05/M	$3.36/M	N/A	N/A	N/A
GLM 5.2 z-ai/glm-5.2:friendli	1.0M	$1.47/M	$4.62/M	$0.27/M	0.3s	107 tps
GLM 5.2 (Thinking) z-ai/glm-5.2:thinking:friendli	1.0M	$1.47/M	$4.62/M	$0.27/M	0.3s	107 tps
GLM 5.3 z-ai/glm-5.3:friendli	1.0M	$1.32/M	$4.16/M	$0.25/M	0.3s	99 tps
GLM 5.3 (Thinking) z-ai/glm-5.3:thinking:friendli	1.0M	$1.32/M	$4.16/M	$0.25/M	0.3s	99 tps
GLM 5.3 Flash z-ai/glm-5.3-flash:friendli	1.0M	$0.16/M	$0.53/M	$0.03/M	1s	129 tps
MiniMax M2.5 minimax/minimax-m2.5:friendli	197k	$0.31/M	$1.26/M	$0.06/M	0.2s	102.5 tps

DeepSeek V3.2

deepseek/deepseek-v3.2:friendli

Context

164k

Input /M

$0.53/M

Output /M

$1.57/M

Cache read /M

$0.26/M

TTFT

2.4s

Throughput

35 tps

Gemma 4 31B

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

Context

262k

Input /M

$0.15/M

Output /M

$0.42/M

Cache read /M

N/A

TTFT

0.6s

Throughput

86 tps

Gemma 4 31B (Thinking)

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

Context

262k

Input /M

$0.15/M

Output /M

$0.42/M

Cache read /M

N/A

TTFT

0.6s

Throughput

86 tps

GLM 5

z-ai/glm-5:friendli

Context

200k

Input /M

$1.05/M

Output /M

$3.36/M

Cache read /M

N/A

TTFT

N/A

Throughput

N/A

GLM 5.2

z-ai/glm-5.2:friendli

Context

1.0M

Input /M

$1.47/M

Output /M

$4.62/M

Cache read /M

$0.27/M

TTFT

0.3s

Throughput

107 tps

GLM 5.2 (Thinking)

z-ai/glm-5.2:thinking:friendli

Context

1.0M

Input /M

$1.47/M

Output /M

$4.62/M

Cache read /M

$0.27/M

TTFT

0.3s

Throughput

107 tps

GLM 5.3

z-ai/glm-5.3:friendli

Context

1.0M

Input /M

$1.32/M

Output /M

$4.16/M

Cache read /M

$0.25/M

TTFT

0.3s

Throughput

99 tps

GLM 5.3 (Thinking)

z-ai/glm-5.3:thinking:friendli

Context

1.0M

Input /M

$1.32/M

Output /M

$4.16/M

Cache read /M

$0.25/M

TTFT

0.3s

Throughput

99 tps

GLM 5.3 Flash

z-ai/glm-5.3-flash:friendli

Context

1.0M

Input /M

$0.16/M

Output /M

$0.53/M

Cache read /M

$0.03/M

TTFT

1s

Throughput

129 tps

MiniMax M2.5

minimax/minimax-m2.5:friendli

Context

197k

Input /M

$0.31/M

Output /M

$1.26/M

Cache read /M

$0.06/M

TTFT

0.2s

Throughput

102.5 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 Friendli via the API

Call a model served by Friendli 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:friendli",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

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