Tinfoil

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 Tinfoil

Model name

14 models

Model	Context	Input /M	Output /M	Cache read /M	Cache hit	TTFT	Throughput
d3 (Decision Large) decision-large	33k	$0.40/M	Free	N/A	N/A	N/A	N/A
d3 Flash (Decision Medium) decision-medium	33k	$0.06/M	Free	N/A	N/A	N/A	N/A
d3 Mini (Decision Small) decision-small	33k	$0.02/M	Free	N/A	N/A	N/A	N/A
DeepSeek V4.1 Flash 1017 TEE TEE verified TEE/deepseek-v4.1-flash:tinfoil	1.0M	$0.682/M	$1.52/M	$0.136/M	56.6%	3.5s	186 tps
Gemma 4 31B TEE verified google/gemma-4-31b-it:tinfoil	262k	$0.42/M	$1.05/M	Supported	0%	5.7s	53.8 tps
Gemma 4 31B (Thinking) TEE verified google/gemma-4-31b-it:thinking:tinfoil	262k	$0.42/M	$1.05/M	Supported	0%	5.7s	53.8 tps
Gemma 4 31B TEE TEE verified TEE/gemma4-31b:tinfoil	262k	$0.42/M	$1.05/M	Supported	N/A	1.7s	83.9 tps
Gemma 4 31B Thinking TEE TEE verified TEE/gemma4-31b:thinking:tinfoil	262k	$0.42/M	$1.05/M	Supported	0%	6s	56.8 tps
GLM 5.3 TEE TEE verified TEE/glm-5.3:tinfoil	1.0M	$1.89/M	$6.04/M	$0.472/M	59.1%	3.3s	128.6 tps
GPT OSS 120B TEE verified openai/gpt-oss-120b:tinfoil	131k	$0.157/M	$0.63/M	Supported	N/A	N/A	N/A
Kimi K3 TEE verified moonshotai/kimi-k3:tinfoil	256k	$4.20/M	$21.00/M	$0.84/M	N/A	N/A	N/A
Kimi K3 TEE TEE verified TEE/kimi-k3:tinfoil	256k	$4.20/M	$21.00/M	$0.84/M	49%	2.5s	90 tps
Llama 3.3 70B TEE/llama3-3-70b	128k	$1.75/M	$2.75/M	$1.75/M	N/A	N/A	N/A
Llama 3.3 70B Instruct TEE verified meta-llama/llama-3.3-70b-instruct:tinfoil	131k	$1.84/M	$2.89/M	Supported	N/A	N/A	N/A

d3 (Decision Large)

decision-large

Context

33k

Input /M

$0.40/M

Output /M

Free

Cache read /M

N/A

Cache hit

N/A

TTFT

N/A

Throughput

N/A

d3 Flash (Decision Medium)

decision-medium

Context

33k

Input /M

$0.06/M

Output /M

Free

Cache read /M

N/A

Cache hit

N/A

TTFT

N/A

Throughput

N/A

d3 Mini (Decision Small)

decision-small

Context

33k

Input /M

$0.02/M

Output /M

Free

Cache read /M

N/A

Cache hit

N/A

TTFT

N/A

Throughput

N/A

DeepSeek V4.1 Flash 1017 TEE

TEE verified

TEE/deepseek-v4.1-flash:tinfoil

Context

1.0M

Input /M

$0.682/M

Output /M

$1.52/M

Cache read /M

$0.136/M

Cache hit

56.6%

TTFT

3.5s

Throughput

186 tps

Gemma 4 31B

TEE verified

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

Context

262k

Input /M

$0.42/M

Output /M

$1.05/M

Cache read /M

Supported

Cache hit

0%

TTFT

5.7s

Throughput

53.8 tps

Gemma 4 31B (Thinking)

TEE verified

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

Context

262k

Input /M

$0.42/M

Output /M

$1.05/M

Cache read /M

Supported

Cache hit

0%

TTFT

5.7s

Throughput

53.8 tps

Gemma 4 31B TEE

TEE verified

TEE/gemma4-31b:tinfoil

Context

262k

Input /M

$0.42/M

Output /M

$1.05/M

Cache read /M

Supported

Cache hit

N/A

TTFT

1.7s

Throughput

83.9 tps

Gemma 4 31B Thinking TEE

TEE verified

TEE/gemma4-31b:thinking:tinfoil

Context

262k

Input /M

$0.42/M

Output /M

$1.05/M

Cache read /M

Supported

Cache hit

0%

TTFT

6s

Throughput

56.8 tps

GLM 5.3 TEE

TEE verified

TEE/glm-5.3:tinfoil

Context

1.0M

Input /M

$1.89/M

Output /M

$6.04/M

Cache read /M

$0.472/M

Cache hit

59.1%

TTFT

3.3s

Throughput

128.6 tps

GPT OSS 120B

TEE verified

openai/gpt-oss-120b:tinfoil

Context

131k

Input /M

$0.157/M

Output /M

$0.63/M

Cache read /M

Supported

Cache hit

N/A

TTFT

N/A

Throughput

N/A

Kimi K3

TEE verified

moonshotai/kimi-k3:tinfoil

Context

256k

Input /M

$4.20/M

Output /M

$21.00/M

Cache read /M

$0.84/M

Cache hit

N/A

TTFT

N/A

Throughput

N/A

Kimi K3 TEE

TEE verified

TEE/kimi-k3:tinfoil

Context

256k

Input /M

$4.20/M

Output /M

$21.00/M

Cache read /M

$0.84/M

Cache hit

49%

TTFT

2.5s

Throughput

90 tps

Llama 3.3 70B

TEE/llama3-3-70b

Context

128k

Input /M

$1.75/M

Output /M

$2.75/M

Cache read /M

$1.75/M

Cache hit

N/A

TTFT

N/A

Throughput

N/A

Llama 3.3 70B Instruct

TEE verified

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

Context

131k

Input /M

$1.84/M

Output /M

$2.89/M

Cache read /M

Supported

Cache hit

N/A

TTFT

N/A

Throughput

N/A

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 Tinfoil via the API

Call a model served by Tinfoil 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": "decision-large",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

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