Starter
Try AI PR review.
- 2 reviewers
- Code Review
- Security Review
- 3 seats included
- 15,000 credits to start, then 5,000/mo
- About 70 reviews to start, then about 25 a month
- Unlimited repos
Flags swallowed errors, unhandled crash paths and dropped logs.
MediumKnown issue in the Log & Errors Review benchmark · caught by DeepSeek V4.1 FlashSee the graded case
IonWarp's models and the review apps, graded on the same open-source pull requests.
| PRs | |||||
|---|---|---|---|---|---|
| IonWarp Log & Errors · Gemini 3.8 Flash | |||||
| Qodo v2 | |||||
| IonWarp Log & Errors · DeepSeek V4.1 Flash | |||||
| Greptile v4.1 | |||||
| Cursor Bugbot | |||||
| IonWarp Log & Errors · Muse Spark 1.3 | |||||
| CodeRabbit | |||||
| IonWarp Log & Errors · GLM 5.3 Flash | |||||
| IonWarp Log & Errors · GPT 5.6 Luna Pro |
On this benchmark Gemini 3.8 Flash scored higher than DeepSeek V4.1 Flash (67.7% F1 against 59.0%). Why IonWarp runs DeepSeek V4.1 Flash: in a trial of our specialist reviewers on production pull requests on 23 September 2026 it filed about as many findings per review as GLM 5.3 Flash at high effort (0.93 against 1.02) for a third of the cost ($0.039 against $0.125), and 3 of its 64 findings were refuted.
One workspace price with seats included. Credits pool across your team. Unlimited repos on every plan.
Log & Errors Review is on Pro and Max, from $49 a month with 5 seats and a 7-day trial.
Try AI PR review.
Get AI PR review right from day 1.
Scale AI PR review.
Custom reviewers + credits · GitHub Enterprise · SSO/SAML + audit logs · Custom security review + SLA
P2wait_produce timer tagged `shards` while its siblings use `shard` · Log & Errors Review
When the flusher produces segments, the two sibling timers/metrics added in this change (`spans.buffer.flusher.produce` at line 185 and `spans.buffer.segment_size_bytes` at line 193) tag the shard dimension with the key `shard`, but `spans.buffer.flusher.wait_produce` tags the exact same dimension with `shards`. Any dashboard, alert or query grouping the flusher timers by the established `shard` tag (the key the neighbours populate) will silently omit wait-produce latency, and the series cannot be joined with `produce`/`segment_size_bytes` for the same shard.