Across 16 comparable benchmarks, Opus 4.5 has better reported results on 13 and worse on 3. 1 further shared result is listed below, each with the reason it is not counted.
Moderate evidence16 of 17 shared results countedHow we compare
Price basis: Standard rates. The price section below shows the applicable periods.
| Fact | GPT OSS 120B | Opus 4.5 |
|---|---|---|
| Context window | 128K tokens | 200K tokens |
| Input / 1M | $0.350 | $5.00 |
| Output / 1M | $0.750 | $25.00 |
| Blended (4:1) | $0.430 | $5.76 |
| Status | Available | Available |
| Released | August 6, 2025 | November 25, 2025 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 7 shared benchmarks grouped by capability on a 0–100 scale. How averages are built
Opus 4.5 ahead on 3 of 4 counted
Opus 4.5 ahead on 1 of 1 counted
No fixed-scale average — see the rows in Benchmark results
GPT OSS 120B ahead on 1 of 1 counted
Opus 4.5 ahead on 2 of 2 counted
Opus 4.5 ahead on 1 of 1 counted
Opus 4.5 ahead on 2 of 3 counted
No shared comparable results for Research & search, Long context, Multimodal, Cybersecurity.
Published scores, with matching modes where available. Gaps show GPT OSS 120B relative to Opus 4.5. Reading the results
| Benchmark | GPT OSS 120B | Opus 4.5 | Gap |
|---|---|---|---|
| APEX-SWERepository Engineering · Pass@1 | 3.3Thinking High · tools | 38.7Thinking High · tools | −35.4 pts#61 vs #36 of 61 |
| LiveCodeBenchAlgorithmic Coding · Pass @K | 87.8Thinking High · no tools | 87.1≠Thinking · no tools | 1.01×#14 vs #17 of 113 |
| SWE-bench VerifiedRepository Engineering · Accuracy | 60.1Thinking · no tools | 80.9≠Extended · tools | −20.8 pts#78 vs #9 of 108 |
| Terminal Bench HardAgentic Development · Accuracy | 23.5Thinking High · tools | 47.0≠Thinking · tools | −23.5 pts#88 vs #15 of 157 |
| SimpleBenchCommonsense · Score (AVG@5) | 22.1Thinking · no tools | 62.0≠Extended · no tools | 0.36×#99 vs #40 of 106 |
| Vending-Bench 2Logical Planning · Money Balance (USD)Not counted · Speed and cost telemetry is reported, not scored | -21.5Unspecified Reasoning · tools | 4967Unspecified Reasoning · tools | —#62 vs #40 of 64 |
| AIME2025Mathematics · Accuracy | 97.9Thinking · tools | 91.3≠Thinking · no tools | +6.6 pts#16 vs #34 of 86 |
| CritPtScientific Reasoning · Score | 1.1Thinking High · no tools | 4.6≠Thinking · no tools | 0.24×#103 vs #73 of 134 |
| GPQA DiamondScientific Reasoning · Accuracy | 80.1Thinking · no tools | 87.0≠Extended · no tools | −6.9 pts#111 vs #70 of 200 |
| Pinch BenchTool Orchestration · Accuracy | 60.6Thinking · tools | 87.2≠Extended · tools | −26.6 pts#36 vs #9 of 38 |
| Creative WritingWriting | 956Normal · no tools | 1686Normal · no tools | 0.57×#104 vs #35 of 121 |
| IF BenchInstruction Following · Accuracy | 69.0Normal · no tools | 58.0≠Extended · tools | +11.0 pts#28 vs #32 of 38 |
| τ³-BankingService Workflows · Score | 12.8Thinking High · tools | 24.7Thinking High · tools | 0.52×#83 vs #48 of 120 |
| HLEKnowledge Exams · Accuracy | 19.0Thinking · tools | 43.2≠Extended · tools | −24.2 pts#107 vs #72 of 151 |
| MMLU-ProKnowledge Exams · Accuracy | 79.0Thinking · no tools | 90.0≠Extended · no tools | −11.0 pts#65 vs #1 of 153 |
| ECICapability Indices · ECI score (capability index, higher is better) | 140Reported best | 150Reported best | 0.93×#97 vs #47 of 168 |
| LiveBenchCross-capability Suites · Accuracy | 46.1Normal · no tools | 72.6≠Thinking · no tools · 64K | −26.5 pts#100 vs #47 of 106 |
Repository Engineering · Pass@1
Algorithmic Coding · Pass @K
Repository Engineering · Accuracy
Agentic Development · Accuracy
Commonsense · Score (AVG@5)
Logical Planning · Money Balance (USD)
Not counted · Speed and cost telemetry is reported, not scored
Mathematics · Accuracy
Scientific Reasoning · Score
Scientific Reasoning · Accuracy
Tool Orchestration · Accuracy
Writing
Instruction Following · Accuracy
Service Workflows · Score
A dash means no result is available.
| Benchmark | GPT OSS 120B | Opus 4.5 |
|---|---|---|
| AIME 2024 | 96.6 | — |
| AIME 2026 | — | 93.3 |
| FrontierMath | — | 20.7 |
| FrontierMath - Tier 4 | — | 4.2 |
| FrontierMath Tier 4 v2 | — | 4.9 |
| FrontierMath v2 | — | 34.4 |
| HLE-Verified | — | 38.8 |
| IMO-ProofBench Advanced | — | 23.8 |
| MathArena Apex | 1.0 | — |
| MMLU | 90.0 | — |
| Riemann-bench | — | 11.2 |
| SciCode | 34.0 | — |
| Aider-Polyglot | 41.8 | — |
| CodeForces | 2622 | — |
| GSO | — | 26.5 |
| SWE-bench | — | 52.6 |
| SWE-bench Multilingual | — | 70.7 |
| SWE-Bench Pro - Commercial | — | 23.4 |
| SWE-Bench Pro - Public | — | 45.9 |
| Terminal Bench 2.0 | — | 59.3 |
| Vibe Code Bench v1.1 | — | 20.6 |
| WeirdML v2 | — | 63.7 |
| WeirdML v3 | — | 0.065 |
| APEX-Agents | 4.4 | — |
| Claw Bench | — | 91.5 |
| MCP-Atlas | — | 69.8 |
| PinchBench v2 | 44.8 | — |
| AA Intelligence Index (historical versions) | 24.1 | — |
| AA Intelligence Index v5.0.1 | 3.8 | — |
| METR Time Horizons v1.1 | — | 293 |
| Text Arena (Coding) | — | 1512 |
| EBR-bench | — | 14.3 |
| LongBench v2 | — | 64.4 |
| GDP.pdf | 4.0 | — |
| GeoBench ACW | — | 75.0 |
| MMMU | — | 80.7 |
| MMMU-Pro | — | 74.0 |
| VPCT | — | 40.0 |
| AutomationBench | 0.15 | — |
| Harvey Lab-AA | 13.9 | — |
| SAGE | — | 52.1 |
| τ²-Bench | — | 82.0 |
| τ²-Bench - Telecom | — | 92.3 |
| ARC-AGI-1 | — | 80.0 |
| ARC-AGI-2 | — | 37.6 |
| BALROG | — | 43.5 |
| MedCode | — | 49.2 |
| MedScribe | — | 85.3 |
Time-of-day providers use peak-hour rates; fixed-price providers use standard rates. Off-peak discounts remain in the published details.
Drag the handle above the bar for cache hits, the one below it for the input : output mix · 5% steps
Blended price = effective input rate at 90% cache hits × 80% + output rate × 20%
GPT OSS 120B costs93%lessthan Opus 4.5 per million tokens — 13.4× cheaper.Using standard rates at this token mix.
GPT OSS 120B: Standard pricing. Cerebras Shared Inference official pricing (gpt-oss-120b) verified 2026-11-13
Opus 4.5: Standard pricing. Anthropic official pricing (platform.claude.com/docs/en/about-claude/pricing) verified 2026-11-13
| Price / 1M tokens | GPT OSS 120B | Opus 4.5 | Ratio |
|---|---|---|---|
| Input | $0.35 | $5.00 | 0.07× |
| Output | $0.75 | $25.00 | 0.03× |
| Cached input | — | $0.50 | — |
| Cache write | — | — | — |
| Blended | $0.43 | $5.76 | 0.07× |
Estimates use the selected token mix. Cache writes, tool fees and long-context surcharges are excluded.
Reported cost per task, paired with the score from the same run. Settings and sources may differ; costs do not affect the verdict.
| Benchmark | GPT OSS 120B | Opus 4.5 |
|---|---|---|
| AA Intelligence Index v5.0.1 | Not reported | |
| ARC-AGI-1 | Not reported | |
| ARC-AGI-2 | Not reported | |
| AutomationBench | $0.01 / task Average · Pass Rate: 0.15 Thinking High (With Tools) · zapier-private-heldout | Not reported |
| WeirdML v3 | Not reported | $8.7034 / task Average · Official score (0-1; higher is better): 0.065 Thinking High (With Tools) |
| Attribute | GPT OSS 120B | Opus 4.5 |
|---|---|---|
| Availability | Available | Available |
| Provider | OpenAI | Anthropic |
| Model type | Reasoning model | Reasoning model |
| First release | August 6, 2025 | November 25, 2025 |
| Context window | 128K tokens | 200K tokens |
| Max output | 128K tokens | 64K tokens |
| Total parameters | 117B | — |
| Active parameters | 5.1B | — |
| Architecture | Mixture of Experts (MoE) | Undisclosed |
| Documented capabilities | Reasoning, Multilingual | Reasoning, Multilingual |
| Knowledge cutoff | — | 2025-03 |
| Input modalities | Text | Text, Image |
| Output modalities | Text | Text |
| Weights | Open weights | Closed / API only |
| Licence | Apache 2.0 | — |
| Commercial use | Commercial use allowed | Proprietary |
These are reported-score comparisons, not controlled head-to-head tests.
What people found comparing these models, or using either one. Comments also appear on the models' own pages.
Used GPT OSS 120B or Opus 4.5? Sign in with Google to share what worked, what did not, and which API or setup you tested.
No community comments on either model yet.
Alternatives with the most benchmark overlap for this matchup.