NextBit

Models served by NextBit

Model	Context	Input /M	Output /M	Cache read /M	Latency	Throughput
DeepSeek V4 Flash FP8 deepseek/deepseek-v4-flash:nextbit	1.0M	$0.15/M	$0.29/M	$0.03/M	2s	67 tps
DeepSeek V4 Flash (Thinking) FP8 deepseek/deepseek-v4-flash:thinking:nextbit	1.0M	$0.15/M	$0.29/M	$0.03/M	2s	67 tps
DeepSeek V4 Flash 0731 FP8 deepseek/deepseek-v4-flash-0731:nextbit	1.0M	$0.46/M	$1.39/M	$0.02/M	9.2s	82 tps
DeepSeek V4 Flash 0731 (Thinking) FP8 deepseek/deepseek-v4-flash-0731:thinking:nextbit	1.0M	$0.46/M	$1.39/M	$0.02/M	9.2s	82 tps
DeepSeek V4 Pro FP8 deepseek/deepseek-v4-pro:nextbit	1.0M	$1.81/M	$3.62/M	$0.15/M	2.7s	43 tps
DeepSeek V4 Pro (Thinking) FP8 deepseek/deepseek-v4-pro:thinking:nextbit	1.0M	$1.81/M	$3.62/M	$0.15/M	2.7s	43 tps
Gemma 4 26B A4B BF16 google/gemma-4-26b-a4b-it:nextbit	262k	$0.10/M	$0.42/M	$0.05/M	0.7s	58 tps
Gemma 4 26B A4B (Thinking) BF16 google/gemma-4-26b-a4b-it:thinking:nextbit	262k	$0.10/M	$0.42/M	$0.05/M	0.7s	58 tps

Prices are per million tokens and include the 5% provider-selection markup. Availability and pricing refresh continuously; each model page shows the live provider comparison.

Using NextBit via the API

Append :nextbit to the model ID, set "provider": "nextbit" in the request body, or send an X-Provider: nextbit header. Explicit provider selection adds a 5% markup over that provider's base price; the prices in the table above already include it.

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:nextbit",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

See the API documentation for provider preferences, price-aware routing, and error behavior.