Hyperfusion

Models served by Hyperfusion

Model name

9 models

Model	Context	Input /M	Output /M	TTFT	Throughput
DeepSeek V4 Flash 0731 FP8 ZDR deepseek/deepseek-v4-flash-0731:hyperfusion	1.0M	$0.0525/M	$0.168/M	N/A	N/A
DeepSeek V4 Flash 0731 (Thinking) FP8 ZDR deepseek/deepseek-v4-flash-0731:thinking:hyperfusion	1.0M	$0.0525/M	$0.168/M	N/A	N/A
Gemma 4 31B BF16 ZDR google/gemma-4-31b-it:hyperfusion	262k	$0.136/M	$0.399/M	N/A	N/A
Gemma 4 31B (Thinking) BF16 ZDR google/gemma-4-31b-it:thinking:hyperfusion	262k	$0.136/M	$0.399/M	N/A	N/A
GPT OSS 120B MXFP4 ZDR openai/gpt-oss-120b:hyperfusion	131k	$0.0315/M	$0.179/M	N/A	N/A
GPT OSS 20B MXFP4 ZDR openai/gpt-oss-20b:hyperfusion	131k	$0.021/M	$0.105/M	N/A	N/A
MiniMax M2.7 FP8 ZDR minimax/minimax-m2.7:hyperfusion	197k	$0.315/M	$1.26/M	N/A	N/A
Qwen3.8 27B BF16 ZDR qwen3.8-27b:hyperfusion	262k	$0.105/M	$1.05/M	N/A	N/A
Qwen3.8 27B Thinking BF16 ZDR qwen3.8-27b:thinking:hyperfusion	262k	$0.105/M	$1.05/M	N/A	N/A

DeepSeek V4 Flash 0731

FP8

ZDR

deepseek/deepseek-v4-flash-0731:hyperfusion

Context

1.0M

Input /M

$0.0525/M

Output /M

$0.168/M

TTFT

N/A

Throughput

N/A

DeepSeek V4 Flash 0731 (Thinking)

FP8

ZDR

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

Context

1.0M

Input /M

$0.0525/M

Output /M

$0.168/M

TTFT

N/A

Throughput

N/A

Gemma 4 31B

BF16

ZDR

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

Context

262k

Input /M

$0.136/M

Output /M

$0.399/M

TTFT

N/A

Throughput

N/A

Gemma 4 31B (Thinking)

BF16

ZDR

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

Context

262k

Input /M

$0.136/M

Output /M

$0.399/M

TTFT

N/A

Throughput

N/A

GPT OSS 120B

MXFP4

ZDR

openai/gpt-oss-120b:hyperfusion

Context

131k

Input /M

$0.0315/M

Output /M

$0.179/M

TTFT

N/A

Throughput

N/A

GPT OSS 20B

MXFP4

ZDR

openai/gpt-oss-20b:hyperfusion

Context

131k

Input /M

$0.021/M

Output /M

$0.105/M

TTFT

N/A

Throughput

N/A

MiniMax M2.7

FP8

ZDR

minimax/minimax-m2.7:hyperfusion

Context

197k

Input /M

$0.315/M

Output /M

$1.26/M

TTFT

N/A

Throughput

N/A

Qwen3.8 27B

BF16

ZDR

qwen3.8-27b:hyperfusion

Context

262k

Input /M

$0.105/M

Output /M

$1.05/M

TTFT

N/A

Throughput

N/A

Qwen3.8 27B Thinking

BF16

ZDR

qwen3.8-27b:thinking:hyperfusion

Context

262k

Input /M

$0.105/M

Output /M

$1.05/M

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 Hyperfusion via the API

Call a model served by Hyperfusion 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-0731:hyperfusion",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

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