Across 17 comparable benchmarks, Claude Sonnet 4.6 has better reported results on 17 and worse on 0. 1 further shared result is listed below, each with the reason it is not counted.
Moderate evidence17 of 18 shared results countedHow we compare
Price basis: Standard rates. The price section below shows the applicable periods.
| Fact | GPT OSS 120B | Claude Sonnet 4.6 |
|---|---|---|
| Context window | 128K tokens | 1M tokens |
| Input / 1M | $0.350 | $3.00 |
| Output / 1M | $0.750 | $15.00 |
| Blended (4:1) | $0.430 | $3.46 |
| Status | Available | Available |
| Released | August 6, 2025 | February 17, 2026 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 9 shared benchmarks grouped by capability on a 0–100 scale. How averages are built
Claude Sonnet 4.6 ahead on 3 of 3 counted
Claude Sonnet 4.6 ahead on 3 of 3 counted
Claude Sonnet 4.6 ahead on 3 of 3 counted
Claude Sonnet 4.6 ahead on 1 of 1 counted
Claude Sonnet 4.6 ahead on 4 of 4 counted
No shared comparable results for Abstract reasoning, Mathematics, Research & search, Long context, Cybersecurity.
Published scores, with matching modes where available. Gaps show GPT OSS 120B relative to Claude Sonnet 4.6. Reading the results
| Benchmark | GPT OSS 120B | Claude Sonnet 4.6 | Gap |
|---|---|---|---|
| APEX-SWERepository Engineering · Pass@1 | 3.3Thinking High · tools | 34.5Thinking High · tools | −31.2 pts#61 vs #43 of 61 |
| SWE-bench VerifiedRepository Engineering · Accuracy | 60.1Thinking · no tools | 79.6Thinking · no tools | −19.5 pts#78 vs #20 of 108 |
| Terminal Bench HardAgentic Development · Accuracy | 23.5Thinking High · tools | 53.0≠Max · tools | −29.5 pts#88 vs #9 of 157 |
| Vending-Bench 2Logical Planning · Money Balance (USD)Not counted · Speed and cost telemetry is reported, not scored | -21.5Unspecified Reasoning · tools | 7204Unspecified Reasoning · tools | —#62 vs #24 of 64 |
| CritPtScientific Reasoning · Score | 1.1Thinking High · no tools | 3.1≠Max · no tools | 0.35×#103 vs #78 of 134 |
| GPQA DiamondScientific Reasoning · Accuracy | 80.1Thinking · no tools | 89.9Thinking · no tools | −9.8 pts#111 vs #52 of 200 |
| SciCodeScientific Computing · Score | 34.0Thinking High · no tools | 50.1≠Max · no tools | 0.68×#88 vs #49 of 98 |
| APEX-AgentsTool Orchestration · Score | 4.4Reported best | 43.0≠Thinking High · tools | 0.10×#61 vs #44 of 61 |
| Pinch BenchTool Orchestration · Accuracy | 60.6Thinking · tools | 88.0Thinking · tools | −27.4 pts#36 vs #6 of 38 |
| PinchBench v2Tool Orchestration · Average score (%) | 44.8Reported best | 62.7Reported best | −17.8 pts#46 vs #37 of 51 |
| GDP.pdfDocuments & Charts · Task pass rate (%) | 4.0Thinking High · no tools | 15.8≠Max · no tools | −11.8 pts#77 vs #40 of 96 |
| Creative WritingWriting | 956Normal · no tools | 1811Normal · no tools | 0.53×#104 vs #23 of 121 |
| AutomationBenchOffice & Business · Pass Rate | 0.15Thinking High · tools· zapier-private-heldout | 5.3≠Max · tools· zapier-private-heldout | −5.2 pts#64 vs #54 of 64 |
| Harvey Lab-AALegal · Criterion pass rate (some entries report all-pass rate) | 13.9Thinking High · tools | 86.0≠Max · tools | −72.1 pts#39 vs #15 of 39 |
| τ³-BankingService Workflows · Score | 12.8Thinking High · tools | 34.4≠Max · tools | 0.37×#83 vs #34 of 120 |
| HLEKnowledge Exams · Accuracy | 19.0Thinking · tools | 49.0Thinking · tools | −30.0 pts#107 vs #53 of 151 |
| ECICapability Indices · ECI score (capability index, higher is better) | 140Reported best | 152Reported best | 0.92×#97 vs #39 of 168 |
| LiveBenchCross-capability Suites · Accuracy | 46.1Normal · no tools | 75.3≠Thinking High · no tools | −29.2 pts#100 vs #32 of 106 |
Repository Engineering · Pass@1
Repository Engineering · Accuracy
Agentic Development · Accuracy
Logical Planning · Money Balance (USD)
Not counted · Speed and cost telemetry is reported, not scored
Scientific Reasoning · Score
Scientific Reasoning · Accuracy
Scientific Computing · Score
Tool Orchestration · Score
Tool Orchestration · Accuracy
Tool Orchestration · Average score (%)
Documents & Charts · Task pass rate (%)
Writing
Office & Business · Pass Rate
Legal · Criterion pass rate (some entries report all-pass rate)
Service Workflows · Score
A dash means no result is available.
| Benchmark | GPT OSS 120B | Claude Sonnet 4.6 |
|---|---|---|
| AIME 2024 | 96.6 | — |
| AIME2025 | 97.9 | — |
| FrontierMath - Tier 4 | — | 8.3 |
| MathArena Apex | 1.0 | — |
| MMLU | 90.0 | — |
| MMLU-Pro | 79.0 | — |
| Aider-Polyglot | 41.8 | — |
| Code Migration | — | 39.9 |
| CodeForces | 2622 | — |
| DeepSWE | — | 29.9 |
| FrontierCode 1.1 | — | 40.0 |
| LiveCodeBench | 87.8 | — |
| Terminal Bench 2.0 | — | 59.1 |
| Terminal-Bench 4.0 | — | 3.0 |
| Vibe Code Bench v1.1 | — | 51.5 |
| MCP-Atlas | — | 69.5 |
| OSWorld 2.0 | — | 41.5 |
| OSWorld-Verified | — | 72.5 |
| ExploitBench | — | 25.8 |
| AA Intelligence Index (historical versions) | 24.1 | — |
| AA Intelligence Index v5.0.1 | 3.8 | — |
| LiveBench 2.0 | — | 65.1 |
| Vals Index | — | 50.6 |
| IF Bench | 69.0 | — |
| Context Arena | — | 82.8 |
| GeoBench ACW | — | 73.0 |
| MMMU-Pro | — | 75.6 |
| AA-AnalystAgent | — | 20.0 |
| EMB (Excel Modeling) | — | 60.1 |
| Finance Agent v2 | — | 51.0 |
| GDPval-AA | — | 57.0 |
| Harvey's Legal Agent Benchmark | — | 5.0 |
| Legal Research Bench | — | 38.5 |
| Public Benefits Bench v1.1 | — | 62.5 |
| SAGE | — | 46.6 |
| Tax Agent Bench | — | 25.8 |
| τ²-Bench - Telecom | — | 97.9 |
| ARC-AGI-1 | — | 86.5 |
| ARC-AGI-2 | — | 60.4 |
| SimpleBench | 22.1 | — |
| BrowseComp | — | 74.7 |
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 costs88%lessthan Claude Sonnet 4.6 per million tokens — 8× 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
Claude Sonnet 4.6: Standard pricing. Anthropic official pricing (platform.claude.com/docs/en/about-claude/pricing) verified 2026-10-21; full 1M context at standard price
| Price / 1M tokens | GPT OSS 120B | Claude Sonnet 4.6 | Ratio |
|---|---|---|---|
| Input | $0.35 | $3.00 | 0.12× |
| Output | $0.75 | $15.00 | 0.05× |
| Cached input | — | $0.30 | — |
| Cache write | — | — | — |
| Blended | $0.43 | $3.46 | 0.12× |
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.
GPT OSS 120B relative to Claude Sonnet 4.6
Summary: 1 of 1 shared benchmarks have matching recorded settings and sources. The average is a geometric mean of 1 positive-cost ratios, one equal weight per benchmark. This benchmark set can include related tests and indexes; it is not a workload cost estimate.
GPT OSS 120B 84× cheaper
GPT OSS 120B 84× cheaper
Effort, protocol, aggregation and source match; other run conditions may differ.
GPT OSS 120B
Claude Sonnet 4.6
| Benchmark | GPT OSS 120B | Claude Sonnet 4.6 |
|---|---|---|
| AA Intelligence Index v5.0.1 | Not reported | |
| ARC-AGI-1 | Not reported | |
| ARC-AGI-2 | Not reported | $2.6966 / task Average · ARC-AGI-2 Score: 60.4 Thinking High (With Tools) |
| DeepSWE | Not reported | |
| FrontierCode 1.1 | Not reported | $2.8979 / task Average · Weighted rubric score: 24.3 Max (With Tools) · frontiercode-1-1-main $2.4401 / task Average · Weighted rubric score: 40.0 Max (With Tools) · frontiercode-1-1-extended $2.2694 / task Average · Weighted rubric score: 39.9 Thinking High (With Tools) · frontiercode-1-1-extended 5 more settings$2.6993 / task Average · Weighted rubric score: 23.5 Thinking High (With Tools) · frontiercode-1-1-main $1.4261 / task Average · Weighted rubric score: 30.4 Thinking Low (With Tools) · frontiercode-1-1-extended $1.6481 / task Average · Weighted rubric score: 15.0 Thinking Low (With Tools) · frontiercode-1-1-main $1.9375 / task Average · Weighted rubric score: 37.5 Thinking Medium (With Tools) · frontiercode-1-1-extended $2.3012 / task Average · Weighted rubric score: 21.1 Thinking Medium (With Tools) · frontiercode-1-1-main |
| LiveBench 2.0 | Not reported | $1.6841 / task Average · Overall score (equal-weight mean of 7 suite scores, 0-100): 65.1 Max (task-specific tools) |
| Tax Agent Bench | Not reported | $1.1935 / task Average · Accuracy (%, rubric × citation quality): 25.8 Unspecified Reasoning (With Tools) |
| Attribute | GPT OSS 120B | Claude Sonnet 4.6 |
|---|---|---|
| Availability | Available | Available |
| Provider | OpenAI | Anthropic |
| Model type | Reasoning model | Chat model |
| First release | August 6, 2025 | February 17, 2026 |
| Context window | 128K tokens | 1M 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 | — | 2026-01 |
| 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 Claude Sonnet 4.6? 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.