DeepSeek V4 Flash (Workers AI) vs GPT-OSS 120B (Workers AI)
GPT-OSS 120B (Workers AI) reads a 26.2K-token pull request for $0.0092, 20% less than DeepSeek V4 Flash (Workers AI) ($0.0115). Both sit in the economy price tier, so the choice between them is about review quality more than budget. The benchmark cannot separate their recall: DeepSeek V4 Flash (Workers AI) minus GPT-OSS 120B (Workers AI) is +8 pts (95% interval −3 pts to +20 pts).
Side by side
| Figure | DeepSeek V4 Flash (Workers AI) | GPT-OSS 120B (Workers AI) |
|---|---|---|
| Price tier | Economy | Economy |
| Bugs caughtshare of known bugs | 49% (better) | 40% |
| False alarmsper clean PR | 1.3 (better) | 2.5 |
| Cost per reviewmeasured | $0.0178 | $0.0075 (better) |
| Cost per bug caught | $0.0309 | $0.0189 (better) |
| Exact linecatches on the expected lines | 88% (better) | 49% |
| Review timemean | 137s | 55s (better) |
| Typical review calllist price | $0.0147 | $0.011 (better) |
| Input / 1M tokens | $0.44 | $0.35 (better) |
| Output / 1M tokens | $1.32 | $0.75 (better) |
| Context window | 1M | 128K |
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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| DeepSeek V4 Flash (Workers AI) | 49% | 46% | 25% | 55% | 46% | 22% | 56% | 61% | 78% | 61% | 44% | 67% | 50% | 0% |
| GPT-OSS 120B (Workers AI) | 40% | 42% | 13% | 46% | 71% | 41% | 11% | 44% | 33% | 78% | 33% | 44% | 33% | 0% |
Of the 42 known bugs both were scored on, both caught 25 and neither caught 8. DeepSeek V4 Flash (Workers AI) caught 5 bugs that GPT-OSS 120B (Workers AI) missed: 1 in Go, 1 in JavaScript, 1 in Kotlin, 1 in PHP and 1 in Swift. GPT-OSS 120B (Workers AI) caught 4 bugs that DeepSeek V4 Flash (Workers AI) missed: 2 in TypeScript, 1 in C# and 1 in Go.
On pull requests with no known bug, DeepSeek V4 Flash (Workers AI) raised 1.3 false alarms per review and GPT-OSS 120B (Workers AI) 2.5. A measured review cost $0.0178 on DeepSeek V4 Flash (Workers AI), 2.4× the cost of $0.0075 on GPT-OSS 120B (Workers AI).
| Pull request | Language · kind | DeepSeek V4 Flash (Workers AI) | GPT-OSS 120B (Workers AI) |
|---|---|---|---|
| nodejs/undicifeat: add configurable maxPayloadSize for WebSocket | JavaScript · Security | 2/3 | 0/3 |
| groue/GRDB.swiftDatabaseQueue and DatabaseSnapshot reads can access task locals | Swift · Concurrency | 1/3 | 0/3 |
| kotest/kotestFix DataClassEq comparator | Kotlin · Logic | 1/3 | 0/3 |
| symfony/symfony[Security] Allow multiple OIDC discovery endpoints | PHP · Security | 1/3 | 0/3 |
| traefik/traefikAdd the option to define custom cipher suites for backend serversTransport | Go · API misuse | 1/4 | 0/6 |
| OrchardCMS/OrchardCoreSimplify using `StringValues` in Liquid templates | C# · Security | 0/3 | 3/3 |
| TanStack/queryfix(query-core): fix hydration bugs for already resolved promises | TypeScript · Logic | 0/4 | 1/6 |
| go-gitea/giteaperf(gitdiff): optimize inline diff highlighting using cache | Go · Error handling | 0/3 | 1/3 |
| trpc/trpcfix(server): pass maxDurationMs timeout signal to subscription handlers | TypeScript · Concurrency | 0/1 | 1/3 |
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: DeepSeek V4 Flash (Workers AI) and GPT-OSS 120B (Workers AI).