coder-v2 on RTX 3090 (24GB) + AMD EPYC 7702P 64-Core Processor (64c) + 252GB
RTX 3090 (24GB) + AMD EPYC 7702P 64-Core Processor (64c) + 252GB
suite suite-v1
cli 0.0.3
signedrJBt+3Kzrz…
Embed badgesubmitted Jul 02, 2026Workload results
| Workload | Backend | Model | decode tok/s | prefill tok/s | TTFT | p50 | p95 |
|---|---|---|---|---|---|---|---|
| chat-short | ollama@0.31.1 | deepseek-coder-v2Q4_0 | 189.5tok/s | 159.2tok/s | 735ms | 5.2ms | 5.6ms |
| chat-long | ollama@0.31.1 | deepseek-coder-v2Q4_0 | 77.56tok/s | 3,907.5tok/s | 838ms | 10.7ms | 11.5ms |
| concurrent-decode | ollama@0.31.1 | deepseek-coder-v2Q4_0 | 155.6tok/s | — | — | 6.3ms | 6.7ms |
| agent-trace | ollama@0.31.1 | deepseek-coder-v2Q4_0 | 101.2tok/s | 6,933.1tok/s | 352ms | 11.3ms | 12.2ms |
Reproduce on your machine
Same workload, same model, signed at your rig. The exact command that produced this run:
$ pipx install llm-speed && llm-speed bench --model 'deepseek-coder-v2' --workload 'chat-short'
Runs in about a minute. Your number lands on the leaderboard signed and linkable. How it's measured.
Embed this run
Drop the badge into a README, blog post, or signature. Each render is a backlink to the signed result.
[](https://llm-speed.com/r/r_o2-1w665rtq)Related benchmarks
- More coder-v2 benchmarks — every backend and rig that has run this model.
- More RTX 3090 (24GB) LLM benchmarks — every model measured on this hardware.
Provenance
- Run ID
- r_o2-1w665rtq
- Fingerprint hash
- Public key
- rJBt+3KzrzeTAA+1BzYrp0h4ONLfSDQFr+RTyzH9Zk0=
- Received
- 2026-07-02 05:14:55