Roadmap
Fintech scale, AI inference, and operating maturity now move on one roadmap.
v0.1 delivered 62,770 TPS on a 3-node DO cluster. v0.2 achieved 436K TPS sharded and 131K TPS VSR. v0.3 shipped April 19 2026 on AWS i4i.4xlarge — 233,894 TPS with live VSR failover testing. v0.3.1 integrated AI inference with Tract ONNX. v0.3.2 added multi-stream priority routing with <1ms critical latency. v0.3.3 hardened the platform with 2PC, mTLS, signed artifacts, and operator runbooks. v0.5 AI LLM verified distributed Qwen2.5-7B generation, and v0.5 AI Production now points toward large-model serving.
v0.1✓ Done62,770 TPS · DO cluster · gRPC
- —Core Rust engine with LMAX Disruptor ring buffer
- —TigerBeetle VSR consensus across three nodes
- —gRPC + Tokio UDP transport · 256 in-flight window
- —3-node DigitalOcean cluster · $252/month · 0% error rate
v0.2✓ Done1.2M TPS local · 436K TPS sharded · 131K TPS VSR
- —Aeron IPC transport — 1,203,108 TPS peak on MacBook Air M4
- —io_uring disk writes · sharded TigerBeetle E2E integration
- —2 shards per node · 436,351 TPS aggregate (3-node sharded)
- —130,998 TPS with full VSR consensus · 2/3 quorum · 0% error
- —DO cluster benchmark completed April 13 2026 · SGP1
v0.3✓ Done233,894 TPS peak · AWS i4i · VSR failover tested
- —Production-grade single-shard with live VSR failover test
- —AWS i4i.4xlarge · Intel Xeon Platinum 8375C · 16 vCPU · 128 GiB · 1.9TB NVMe
- —4 shards × dedicated TigerBeetle client · 3-node VSR cluster (loopback)
- —233,894 TPS peak · 180,500 TPS avg · 12,421,068 events · 0% error
- —VSR failover tested: 1 replica killed at t=80s → recovered in 37s · bench continued
- —AWS Singapore · April 19 2026
v0.3 — AWS i4i.4xlarge · 233,894 TPS peak · VSR failover recovery tested · 12,421,068 events · 0% error · April 19 2026
v0.3.1 AI🔨 Implemented1,500–2,000 RPS estimate · AI inference pipeline
- —5 production-grade datasets implemented
- —Tract ONNX runtime integration for AI inference
- —io_uring dataloader for high-throughput model serving
- —2,291 LOC · 57 tests passing · CI 100% green
- —AI inference pipeline implemented · production benchmark pending
v0.3.2 Priority✓ DoneSame TPS · <1ms critical latency
- —Multi-stream priority routing (Critical/High/Normal)
- —Critical requests bypass queue with <1ms latency guarantee
- —429 tests passing · 0 Clippy warnings
- —Production-ready priority scheduler deployed
v0.3.3 Hardening✓ DoneSame TPS · 2PC · mTLS · signed supply chain
- —Cross-shard 2PC via TigerBeetle pending/post/void flow
- —Kubernetes ingress + cert-manager mTLS automation
- —Prometheus PVC persistence and operational hardening
- —Syft SBOM + Cosign keyless signing in CI
- —3 ADRs and 8 runbooks published for operators
v0.4 Fintech Scale🔮 Future1M+ TPS target · 4× i8g.16xlarge Graviton 4
- —4× AWS i8g.16xlarge instances (Graviton 4, 64 vCPU each)
- —Sharded VSR cluster with horizontal scaling
- —Production benchmark pending · target 1M+ sustained TPS
- —Multi-region replication architecture
Target: 2027 — bare-metal NVMe Gen4 · XDP ingress · estimated 5–10M TPS sharded · 1–2M TPS VSR
v0.5 AI LLM✓ Verified32 tokens in 19.7s · distributed Qwen2.5-7B
- —Qwen2.5-7B-Instruct distributed 3-stage inference pipeline
- —Stage orchestration across Aeron IPC streams 1001/2001/2002/1002/1003
- —Language Drift issue fixed and English generation verified
- —KV cache preserved across decode steps with correct position propagation
- —Production-ready multi-token generation on CPU-only Apple M4
Qwen2.5-7B-Instruct · 3-stage distributed pipeline · 32 tokens in 19.7s on Apple M4 CPU · multi-token generation verified
v0.5 AI Production🔮 FutureTBD RPS · AI inference production benchmark
- —Cortex v1 on ClarkenAI 70B plus Blazil Super Engine
- —Cloud cost validation against commercial inference APIs
- —Ankatos runtime integration and shared transport layer
- —Production benchmark pending for large-model serving
The source is open and the benchmarks are reproducible.
Clone the repository, run the benchmark suite, and verify every number yourself.