Products/Developer Tools/tokenspeedQuietVC Report ↗

tokenspeed

TokenSpeed is a lightweight Python-based inference engine designed to maximize throughput and minimize latency for large language models

Developer Tools·Launched May 2026·0 upvotes·lightseek.org

▲Upvote0 Visit Website↗

ALIVEverified Jul 16, 2026 · HTTP 200

CohortClass of 2026

What it is

TokenSpeed is a lightweight Python-based inference engine designed to maximize throughput and minimize latency for large language models. It optimizes token generation speed across popular LLM architectures including Deepseek, Qwen, and Kimi, enabling faster real-time inference at scale.

HTTP liveness

23 probes since May 23

May 23Jul 16

AliveHard-dead (404/5xx/parked)RedirectedUnverified (timeout/DNS/block)

Observed events

probes · press · GitHub

Jul 4, 2026
+95 stars
1,543 total
Jun 17, 2026
+203 stars
1,448 total
May 28, 2026
+48 stars
1,245 total
May 27, 2026
+17 stars
1,197 total
May 26, 2026
+91 stars
1,180 total
May 24, 2026
+17 stars
1,089 total
May 23, 2026
First probe — live
Alive · 200 OK
May 21, 2026
+24 stars
1,072 total

Funding history

1 round

Total raised

$2.0B

Latest valuation

$20.0B

Stage

unknown

Lead investors: Meituan

Private

Apr 2026

Meituan

Valuation: $20.0B

$2.0B

source

Data from SEC EDGAR Form D filings and press coverage. Round labels for SEC filings are inferred from filing sequence and amount.

Moat

None apparent

Optimizes token generation speed across popular LLM architectures.

moderate · 0.80 confidence

Text classification · Jun 2026

Discussion 0 comments

No comments yet. Be the first!

VitalsQuiet

Upvotes

—

Press

—

Command Palette