GLM-5.3 (Workers AI) vs GPT-OSS 120B (Workers AI)
Reading a 26.2K-token pull request costs $0.0367 on GLM-5.3 (Workers AI), 4.0× the $0.0092 it costs on GPT-OSS 120B (Workers AI). GLM-5.3 (Workers AI) is a mid-range model and GPT-OSS 120B (Workers AI) a economy one. The benchmark cannot separate their recall: GLM-5.3 (Workers AI) minus GPT-OSS 120B (Workers AI) is +21 pts (95% interval +10 pts to +32 pts).
Side by side
| Figure | GLM-5.3 (Workers AI) | GPT-OSS 120B (Workers AI) |
|---|---|---|
| Price tier | Mid-range | Economy |
| Bugs caughtshare of known bugs | 61% (better) | 40% |
| False alarmsper clean PR | 1.9 (better) | 2.5 |
| Cost per reviewmeasured | $0.0311 | $0.0075 (better) |
| Cost per bug caught | $0.0559 | $0.0189 (better) |
| Exact linecatches on the expected lines | 85% (better) | 49% |
| Review timemean | 65s | 55s (better) |
| Typical review calllist price | $0.0473 | $0.011 (better) |
| Input / 1M tokens | $1.40 | $0.35 (better) |
| Output / 1M tokens | $4.40 | $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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| GLM-5.3 (Workers AI) | 61% | 78% | 19% | 46% | 73% | 48% | 33% | 72% | 44% | 58% | 67% | 100% | 100% | 67% |
| 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 28 and neither caught 5. GLM-5.3 (Workers AI) caught 8 bugs that GPT-OSS 120B (Workers AI) missed: 1 in C, 1 in C++, 1 in Go, 1 in JavaScript, 1 in Kotlin, 1 in Rust, 1 in Swift and 1 in TypeScript. GPT-OSS 120B (Workers AI) caught one bug that GLM-5.3 (Workers AI) missed: 1 in Go.
On pull requests with no known bug, GLM-5.3 (Workers AI) raised 1.9 false alarms per review and GPT-OSS 120B (Workers AI) 2.5. A measured review cost $0.0311 on GLM-5.3 (Workers AI), 4.1× the cost of $0.0075 on GPT-OSS 120B (Workers AI).
| Pull request | Language · kind | GLM-5.3 (Workers AI) | GPT-OSS 120B (Workers AI) |
|---|---|---|---|
| ggml-org/llama.cpprpc : cache and reuse compute graphs | C++ · Security | 3/3 | 0/3 |
| groue/GRDB.swiftDatabaseQueue and DatabaseSnapshot reads can access task locals | Swift · Concurrency | 3/3 | 0/3 |
| 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 | 2/3 | 0/3 |
| kotest/kotestFix DataClassEq comparator | Kotlin · Logic | 2/3 | 0/3 |
| nodejs/undicifeat: add configurable maxPayloadSize for WebSocket | JavaScript · Security | 2/3 | 0/3 |
| traefik/traefikAdd the option to define custom cipher suites for backend serversTransport | Go · API misuse | 2/4 | 0/6 |
| colinhacks/zodfeat(v4): parse input containing reference cycles | TypeScript · Resource handling | 1/4 | 0/6 |
| nushell/nushellreorganize nushell configuration | Rust · Error handling | 1/4 | 0/6 |
| go-gitea/giteaperf(gitdiff): optimize inline diff highlighting using cache | Go · Error handling | 0/3 | 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: GLM-5.3 (Workers AI) and GPT-OSS 120B (Workers AI).