Across 18 comparable benchmarks, Haiku 4.5 has better reported results on 11 and worse on 7. 1 further shared result is listed below, each with the reason it is not counted.
Moderate evidence18 of 19 shared results countedHow we compare
Price basis: Standard rates. The price section below shows the applicable periods.
| Fact | Haiku 4.5 | GPT OSS 120B |
|---|---|---|
| Context window | 200K tokens | 128K tokens |
| Input / 1M | $1.00 | $0.350 |
| Output / 1M | $5.00 | $0.750 |
| Blended (4:1) | $1.15 | $0.430 |
| Status | Available | Available |
| Released | October 15, 2025 | August 6, 2025 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 10 shared benchmarks grouped by capability on a 0–100 scale. How averages are built
Haiku 4.5 ahead on 2 of 3 counted
GPT OSS 120B ahead on 1 of 1 counted
Split 1–1 of 2 counted
Haiku 4.5 ahead on 2 of 2 counted
GPT OSS 120B ahead on 1 of 1 counted
Split 2–2 of 4 counted
No shared comparable results for Abstract reasoning, Research & search, Long context, Cybersecurity.
Published scores, with matching modes where available. Gaps show Haiku 4.5 relative to GPT OSS 120B. Reading the results
| Benchmark | Haiku 4.5 | GPT OSS 120B | Gap |
|---|---|---|---|
| LiveCodeBenchAlgorithmic Coding · Pass @K | 62.0Extended · no tools | 87.8≠Thinking High · no tools | 0.71×#69 vs #14 of 113 |
| SWE-bench VerifiedRepository Engineering · Accuracy | 73.3Thinking · tools · 128K | 60.1≠Thinking · no tools | +13.2 pts#47 vs #78 of 108 |
| Terminal Bench HardAgentic Development · Accuracy | 27.3Normal · tools | 23.5≠Thinking High · tools | +3.8 pts#74 vs #88 of 157 |
| Vending-Bench 2Logical Planning · Money Balance (USD)Not counted · Speed and cost telemetry is reported, not scored | 459Unspecified Reasoning · tools | -21.5Unspecified Reasoning · tools | —#58 vs #62 of 64 |
| AIME2025Mathematics · Accuracy | 96.3Thinking · tools · 128K | 97.9≠Thinking · tools | −1.6 pts#18 vs #16 of 86 |
| GPQA DiamondScientific Reasoning · Accuracy | 73.3Extended · no tools | 80.1≠Thinking · no tools | −6.8 pts#126 vs #111 of 200 |
| SciCodeScientific Computing · Score | 42.2Thinking · no tools | 34.0≠Thinking High · no tools | 1.24×#72 vs #88 of 98 |
| Pinch BenchTool Orchestration · Accuracy | 82.0Thinking · tools | 60.6Thinking · tools | +21.4 pts#22 vs #36 of 38 |
| PinchBench v2Tool Orchestration · Average score (%) | 67.7Reported best | 44.8Reported best | +22.9 pts#33 vs #46 of 51 |
| GDP.pdfDocuments & Charts · Task pass rate (%) | 3.8Thinking · no tools | 4.0≠Thinking High · no tools | −0.2 pts#78 vs #77 of 96 |
| IF BenchInstruction Following · Accuracy | 54.3Extended · no tools | 69.0≠Normal · no tools | −14.7 pts#36 vs #28 of 38 |
| AutomationBenchOffice & Business · Pass Rate | 0.46Reported best· zapier-private-heldout | 0.15≠Thinking High · tools· zapier-private-heldout | +0.3 pts#62 vs #64 of 64 |
| Harvey Lab-AALegal · Criterion pass rate (some entries report all-pass rate) | 61.0Thinking · tools | 13.9≠Thinking High · tools | +47.2 pts#31 vs #39 of 39 |
| τ³-BankingService Workflows · Score | 9.3Thinking · tools | 12.8≠Thinking High · tools | 0.73×#95 vs #83 of 120 |
| HLEKnowledge Exams · Accuracy | 9.7Extended · no tools | 19.0≠Thinking · tools | −9.3 pts#129 vs #107 of 151 |
| MMLU-ProKnowledge Exams · Accuracy | 80.0Normal · no tools | 79.0≠Thinking · no tools | +1.0 pts#62 vs #65 of 153 |
| AA Intelligence Index v5.0.1Capability Indices · Index (0-100) | 7.4Thinking · tools | 3.8≠Thinking High · tools | 1.95×#54 vs #62 of 62 |
| ECICapability Indices · ECI score (capability index, higher is better) | 142Reported best | 140Reported best | 1.02×#90 vs #97 of 168 |
| LiveBenchCross-capability Suites · Accuracy | 61.3Thinking · no tools · 64K | 46.1≠Normal · no tools | +15.2 pts#82 vs #100 of 106 |
Algorithmic Coding · Pass @K
Repository Engineering · Accuracy
Agentic Development · Accuracy
Logical Planning · Money Balance (USD)
Not counted · Speed and cost telemetry is reported, not scored
Mathematics · Accuracy
Scientific Reasoning · Accuracy
Scientific Computing · Score
Tool Orchestration · Accuracy
Tool Orchestration · Average score (%)
Documents & Charts · Task pass rate (%)
Instruction Following · Accuracy
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 | Haiku 4.5 | GPT OSS 120B |
|---|---|---|
| AIME 2024 | — | 96.6 |
| CritPt | — | 1.1 |
| FrontierMath | 4.1 | — |
| FrontierMath - Tier 4 | 2.1 | — |
| MathArena Apex | — | 1.0 |
| MMLU | — | 90.0 |
| Aider-Polyglot | — | 41.8 |
| APEX-SWE | — | 3.3 |
| Code Migration | 7.5 | — |
| CodeForces | — | 2622 |
| SWE-Bench Pro - Public | 39.5 | — |
| SWE-Bench Pro V2 Hard | 25.5 | — |
| Terminal-Bench 1.0 | 41.8 | — |
| Vibe Code Bench v1.1 | 11.4 | — |
| APEX-Agents | — | 4.4 |
| Claw Bench | 89.4 | — |
| MCP-Atlas | 40.2 | — |
| OSWorld 2.1 | 15.7 | — |
| ExploitBench | 13.7 | — |
| AA Intelligence Index (historical versions) | — | 24.1 |
| LiveBench 2.0 | 46.0 | — |
| Vals Index | 22.9 | — |
| Creative Writing | — | 956 |
| Context Arena | 36.3 | — |
| MMMU | 73.2 | — |
| MMMU-Pro | 58.6 | — |
| AA-AnalystAgent | 15.0 | — |
| EMB (Excel Modeling) | 22.7 | — |
| Finance Agent v2 | 31.0 | — |
| Harvey's Legal Agent Benchmark | 0.83 | — |
| Legal Research Bench | 10.6 | — |
| Public Benefits Bench v1.1 | 54.3 | — |
| SAGE | 31.8 | — |
| τ²-Bench | 33.0 | — |
| ARC-AGI-1 | 47.7 | — |
| ARC-AGI-2 | 4.5 | — |
| BALROG | 31.2 | — |
| SimpleBench | — | 22.1 |
| GAIA | 56.4 | — |
| MedCode | 32.7 | — |
| MedScribe | 85.2 | — |
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%
Haiku 4.5 costs168%morethan GPT OSS 120B per million tokens — 2.7× the price.Using standard rates at this token mix.
Haiku 4.5: Standard pricing. Anthropic official pricing (platform.claude.com/docs/en/about-claude/pricing) verified 2026-11-13
GPT OSS 120B: Standard pricing. Cerebras Shared Inference official pricing (gpt-oss-120b) verified 2026-11-13
| Price / 1M tokens | Haiku 4.5 | GPT OSS 120B | Ratio |
|---|---|---|---|
| Input | $1.00 | $0.35 | 2.86× |
| Output | $5.00 | $0.75 | 6.67× |
| Cached input | $0.10 | — | — |
| Cache write | — | — | — |
| Blended | $1.15 | $0.43 | 2.68× |
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.
Haiku 4.5 relative to GPT OSS 120B
No matching recorded settings for a cost summary.
Summary: 0 of 2 shared benchmarks have matching recorded settings and sources. The average is a geometric mean of 0 positive-cost ratios, one equal weight per benchmark. This benchmark set can include related tests and indexes; it is not a workload cost estimate.
Settings or sources differ
Settings or sources differ
Settings or sources differ
Settings or sources differ
Settings or sources differ or are not recorded; interpret the comparison with care.
Haiku 4.5
GPT OSS 120B
Settings or sources differ or are not recorded; interpret the comparison with care.
Haiku 4.5
GPT OSS 120B
| Benchmark | Haiku 4.5 | GPT OSS 120B |
|---|---|---|
| ARC-AGI-1 | Not reported | |
| LiveBench 2.0 | $0.934 / task Average · Overall score (equal-weight mean of 7 suite scores, 0-100): 46.0 Max (task-specific tools) | Not reported |
| Public Benefits Bench v1.1 | $0.2245 / task Average · Criterion pass rate (%): 54.3 Unspecified Reasoning (With Tools) | Not reported |
| Attribute | Haiku 4.5 | GPT OSS 120B |
|---|---|---|
| Availability | Available | Available |
| Provider | Anthropic | OpenAI |
| Model type | Multimodal model | Reasoning model |
| First release | October 15, 2025 | August 6, 2025 |
| Context window | 200K tokens | 128K tokens |
| Max output | 64K tokens | 128K tokens |
| Total parameters | — | 117B |
| Active parameters | — | 5.1B |
| Architecture | Undisclosed | Mixture of Experts (MoE) |
| Documented capabilities | Reasoning, Multilingual | Reasoning, Multilingual |
| Input modalities | Text, Image | Text |
| Output modalities | Text | Text |
| Weights | Closed / API only | Open weights |
| Licence | — | Apache 2.0 |
| Commercial use | Proprietary | Commercial use allowed |
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 Haiku 4.5 or GPT OSS 120B? 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.