Claude Haiku 5.5 vs Claude Opus 5.5
Reading a 26.2K-token pull request costs $0.1048 on Claude Opus 5.5, 40× the $0.0026 it costs on Claude Haiku 5.5. Claude Haiku 5.5 is a economy model and Claude Opus 5.5 a frontier one. The benchmark cannot separate their recall: Claude Haiku 5.5 minus Claude Opus 5.5 is −26 pts (95% interval −40 pts to −12 pts).
Side by side
| Figure | Claude Haiku 5.5 | Claude Opus 5.5 |
|---|---|---|
| Price tier | Economy | Frontier |
| Bugs caughtshare of known bugs | 64% | 90% (better) |
| False alarmsper clean PR | 1.4 (better) | 1.7 |
| Cost per reviewmeasured | $0.0038 (better) | $0.1525 |
| Cost per bug caught | $0.0064 (better) | $0.1921 |
| Exact linecatches on the expected lines | 82% | 89% (better) |
| Review timemean | 15s (better) | 33s |
| Typical review calllist price | $0.0038 (better) | $0.1532 |
| Input / 1M tokens | $0.10 (better) | $4.00 |
| Output / 1M tokens | $0.50 (better) | $20.00 |
| Context window | 1M | 1M |
Head to head
The known bugs both were scored on, by who caught each in at least one run.
By language
| Model | All | TypeScript6 bugs | Go4 bugs | Java4 bugs | Rust4 bugs | C#3 bugs | JavaScript3 bugs | Kotlin3 bugs | PHP3 bugs | Python3 bugs | Ruby3 bugs | Swift3 bugs | C++2 bugs | C1 bug |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Claude Haiku 5.5 | 64% | 83% | 75% | 40% | 75% | 67% | 33% | 67% | 100% | 39% | 67% | 67% | 50% | 0% |
| Claude Opus 5.5 | 90% | 100% | 50% | 78% | 100% | 100% | 100% | 67% | 100% | 100% | 89% | 100% | 100% | 100% |
Of the 42 known bugs both were scored on, both caught 27 and neither caught 2. Claude Haiku 5.5 caught one bug that Claude Opus 5.5 missed: 1 in Go. Claude Opus 5.5 caught 12 bugs that Claude Haiku 5.5 missed: 2 in Java, 2 in JavaScript, 1 in C, 1 in C++, 1 in C#, 1 in Python, 1 in Ruby, 1 in Rust, 1 in Swift and 1 in TypeScript.
On pull requests with no known bug, Claude Haiku 5.5 raised 1.4 false alarms per review and Claude Opus 5.5 1.7. A measured review cost $0.0038 on Claude Haiku 5.5, 97% less than $0.1525 on Claude Opus 5.5.
| Pull request | Language · kind | Claude Haiku 5.5 | Claude Opus 5.5 |
|---|---|---|---|
| go-gitea/giteaperf(gitdiff): optimize inline diff highlighting using cache | Go · Error handling | 1/1 | 0/1 |
| TanStack/queryfix(query-core): fix hydration bugs for already resolved promises | TypeScript · Logic | 0/1 | 1/1 |
| aio-libs/aiohttpkeep websocket compression state across interleaved control frames | Python · Logic | 0/1 | 1/1 |
| c-ares/c-aresEvent Thread could stall when not notified of new queries on existing connections that are in a bad state | C · Resource handling | 0/1 | 1/1 |
| dotnet/orleansfix(serialization): cap codec allocations | C# · Logic | 0/1 | 1/1 |
| ggml-org/llama.cpprpc : cache and reuse compute graphs | C++ · Security | 0/1 | 1/1 |
| jhy/jsoupStore cached child els in Attributes, not field | Java · Concurrency | 0/2 | 1/2 |
| mockito/mockitoAdd core API to enable Kotlin singleton mocking | Java · API misuse | 0/1 | 1/1 |
| nushell/nushellreorganize nushell configuration | Rust · Error handling | 0/1 | 1/1 |
| pinojs/thread-streamrefactor: replace setTimeout with Atomics.waitAsync | JavaScript · Concurrency | 0/1 | 1/1 |
| pointfreeco/swift-navigationUIKit: Fix switch presented child double dismiss | Swift · Logic | 0/1 | 1/1 |
| rails/railsReimplement Action Cable redis adapter with redis-client | Ruby · Logic | 0/1 | 2/3 |
| sindresorhus/p-throttleAdd `weight` option | JavaScript · Logic | 0/1 | 1/1 |
Defects caught over all runs: a PR reviewed three times with one known bug shows 3/3 when it was caught every time.
Every bug each was scored on: Claude Haiku 5.5 and Claude Opus 5.5.