GPT-5.6 Sol vs GPT-OSS 120B (Workers AI)
Reading a 26.2K-token pull request costs $0.1048 on GPT-5.6 Sol, 11× the $0.0092 it costs on GPT-OSS 120B (Workers AI). GPT-5.6 Sol is a frontier model and GPT-OSS 120B (Workers AI) a economy one. In the benchmark GPT-5.6 Sol caught 32 percentage points more of the known defects than GPT-OSS 120B (Workers AI) (95% interval +18 pts to +46 pts, a − b).
Side by side
| Figure | GPT-5.6 Sol | GPT-OSS 120B (Workers AI) |
|---|---|---|
| Price tier | Frontier | Economy |
| Bugs caughtshare of known bugs | 72% (better) | 40% |
| False alarmsper clean PR | 1.2 (better) | 2.5 |
| Cost per reviewmeasured | $0.1127 | $0.0075 (better) |
| Cost per bug caught | $0.1732 | $0.0189 (better) |
| Exact linecatches on the expected lines | 92% (better) | 49% |
| Review timemean | 50s (better) | 55s |
| Typical review calllist price | $0.1532 | $0.011 (better) |
| Input / 1M tokens | $4.00 | $0.35 (better) |
| Output / 1M tokens | $20.00 | $0.75 (better) |
| Context window | 1.1M | 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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| GPT-5.6 Sol | 72% | 67% | 50% | 69% | 75% | 56% | 100% | 67% | 100% | 100% | 67% | 100% | 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 26 and neither caught 7. GPT-5.6 Sol caught 6 bugs that GPT-OSS 120B (Workers AI) missed: 2 in JavaScript, 1 in Go, 1 in PHP, 1 in Swift and 1 in TypeScript. GPT-OSS 120B (Workers AI) caught 3 bugs that GPT-5.6 Sol missed: 2 in TypeScript and 1 in Go.
On pull requests with no known bug, GPT-5.6 Sol raised 1.2 false alarms per review and GPT-OSS 120B (Workers AI) 2.5. A measured review cost $0.1127 on GPT-5.6 Sol, 15× the cost of $0.0075 on GPT-OSS 120B (Workers AI).
| Pull request | Language · kind | GPT-5.6 Sol | GPT-OSS 120B (Workers AI) |
|---|---|---|---|
| colinhacks/zodfeat(v4): parse input containing reference cycles | TypeScript · Resource handling | 1/1 | 0/6 |
| go-chi/chifeat: middleware.ClientIP, a replacement for middleware.RealIP | Go · Security | 1/1 | 0/3 |
| groue/GRDB.swiftDatabaseQueue and DatabaseSnapshot reads can access task locals | Swift · Concurrency | 1/1 | 0/3 |
| nodejs/undicifeat: add configurable maxPayloadSize for WebSocket | JavaScript · Security | 1/1 | 0/3 |
| sindresorhus/p-throttleAdd `weight` option | JavaScript · Logic | 1/1 | 0/3 |
| symfony/symfony[Security] Allow multiple OIDC discovery endpoints | PHP · Security | 1/1 | 0/3 |
| go-gitea/giteaperf(gitdiff): optimize inline diff highlighting using cache | Go · Error handling | 0/1 | 1/3 |
| trpc/trpcfix(server): pass maxDurationMs timeout signal to subscription handlers | TypeScript · Concurrency | 0/1 | 1/3 |
| withastro/astroPort Go's csrf protection | TypeScript · Logic | 0/1 | 5/6 |
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: GPT-5.6 Sol and GPT-OSS 120B (Workers AI).