Fireworks AI DeepSeek V3

Fireworks AI DeepSeek V3: $0.9/1M input tokens, 420 tok/s throughput, 380ms latency. Compare on BenchNode.io.

$0.9

per 1M input tokens

$0.9 / 1M output

Free tier available

Specifications

model id: deepseek-v3
model family: DeepSeek V3
parameters: 685B MoE
context window tokens: 128,000
modality: text
reasoning: No

Performance

Latency (TTFT) 380 ms

lower is better

Uptime SLA 99.9%

Throughput 420 tok/s

380

ms latency

99.9%

uptime

Pricing Detail

input per 1m tokens usd: 0.9
output per 1m tokens usd: 0.9
free tier available: Yes
rate limit tpm: 10,000
rate limit rpm: 60

AI Analysis · gpt-4o-mini

Technical Verdict

The input and output costs are $0.9 per 1M tokens, which is mid-tier for this model class, while the throughput of 420 tok/s is relatively fast. The rate limits of 10,000 TPM and 60 RPM suggest suitability for low-volume prototyping rather than high-volume production, and the first-token latency of 380 ms leans towards batch use over real-time applications.