Model comparison
Claude 3 Haiku vs Claude Opus 4.5
Head-to-head evidence from 12 shared benchmark results across 6 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: Claude 3 Haiku #194 (Supported); Claude Opus 4.5 #34 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. Claude 3 Haiku and Claude Opus 4.5 share 12 comparable benchmark results. 0 of 8 categories are comparable. 0 results are unique to Claude 3 Haiku; 47 to Claude Opus 4.5.
Updated July 22, 2026- Shared results
- 12
- Claude 3 Haiku only
- 0
- Claude Opus 4.5 only
- 47
- Comparable categories
- 0 / 8
Benchmark data for Claude 3 Haiku and Claude Opus 4.5 is coming soon on BenchLM.
Confidence note. This is a partial-evidence comparison with 12 shared benchmark results across 6 evidence categories; 0 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
BenchLM has partial data for these models, but not enough overlapping benchmark coverage to produce a fair score-level comparison yet.
Claude Opus 4.5 is priced at $5.00 input / $25.00 output per 1M tokens, versus $0.25 input / $1.25 output per 1M tokens for Claude 3 Haiku.
Category breakdown
Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.
| Category | Claude 3 Haiku | Δ | Claude Opus 4.5 |
|---|---|---|---|
| Agentic | Claude 3 HaikuNot measured | MarginNo overlap | Claude Opus 4.562.6 |
| Coding | Claude 3 HaikuNot measured | MarginNo overlap | Claude Opus 4.571.7 |
| Reasoning | Claude 3 HaikuNot measured | MarginNo overlap | Claude Opus 4.564.4 |
| Knowledge | Claude 3 HaikuNot measured | MarginNo overlap | Claude Opus 4.558.1 |
| Math | Claude 3 HaikuNot measured | MarginNo overlap | Claude Opus 4.557.5 |
| Multilingual | Claude 3 HaikuNot measured | MarginNo overlap | Claude Opus 4.585.7 |
| Multimodal | Claude 3 HaikuNot measured | MarginNo overlap | Claude Opus 4.569.9 |
| Inst. Following | Claude 3 HaikuNot measured | MarginNo overlap | Claude Opus 4.569.5 |
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | Claude 3 Haiku | Claude Opus 4.5 | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Claude 3 Haiku$0.25 input / $1.25 output | Claude Opus 4.5$5 input / $25 output | Claude 3 Haiku has the lower combined listed price. |
| Generation speedtokens per second | Claude 3 Haiku138 tok/s | Claude Opus 4.546 tok/s | Claude 3 Haiku has the higher measured throughput. |
| First-answer latencyseconds to first token | Claude 3 Haiku1.16 s | Claude Opus 4.51.01 s | Claude Opus 4.5 reaches the first token sooner. |
| Context windowmaximum listed tokens | Claude 3 Haiku200K | Claude Opus 4.5200K | Listed context windows are equal. |
Benchmark Deep Dive
Agentic16 benchmarks
| Benchmark | Claude 3 Haiku | Claude Opus 4.5 | Result |
|---|---|---|---|
| τ²-bench resultsSource | 21.1% | 86.3% | Claude Opus 4.5 leads |
| Terminal-Bench 2.0Source | — | 59.3% | Not comparable |
| OSWorld-VerifiedSource | — | 66.3% | Not comparable |
| OSWorldSource | — | 66.3% | Not comparable |
| Claw-EvalSource | — | 59.6% | Not comparable |
| QwenClawBenchSource | — | 52.3% | Not comparable |
| τ³-bench resultsSource | — | 70.2% | Not comparable |
| VITA-BenchSource | — | 23.3% | Not comparable |
| DeepPlanningSource | — | 26.4% | Not comparable |
| ToolathlonSource | — | 43.5% | Not comparable |
| MCP AtlasSource | — | 42.3% | Not comparable |
| MCP-TasksSource | — | 71.8% | Not comparable |
| WideResearchSource | — | 76.4% | Not comparable |
| CyberGymSource | — | 50.6% | Not comparable |
| Gert LabsSource | — | 64.23% | Not comparable |
| JobBenchSource | — | 32.3% | Not comparable |
Coding6 benchmarks
Reasoning4 benchmarks
Knowledge13 benchmarks
| Benchmark | Claude 3 Haiku | Claude Opus 4.5 | Result |
|---|---|---|---|
| Artificial Analysis Intelligence IndexSource | 3.9% | 34.7% | Claude Opus 4.5 leads |
| AA-GPQA DiamondSource | 37.4% | 81.0% | Claude Opus 4.5 leads |
| AA-HLESource | 3.9% | 12.9% | Claude Opus 4.5 leads |
| AA-Omniscience IndexSource | -47.6% | -3.9% | Claude Opus 4.5 leads |
| AA-Omniscience AccuracySource | 17.2% | 40.7% | Claude Opus 4.5 leads |
| AA-Omniscience Hallucination RateSource | 78.2% | 75.4% | Claude Opus 4.5 leads |
| GPQASource | — | 87% | Not comparable |
| SuperGPQASource | — | 70.6% | Not comparable |
| MMLU-ProSource | — | 89.5% | Not comparable |
| MMLU-ReduxSource | — | 96.6% | Not comparable |
| C-EvalSource | — | 92.2% | Not comparable |
| HLESource | — | 30.8% | Not comparable |
| AA MMLU-ProSource | — | 88.9% | Not comparable |
Math7 benchmarks
| Benchmark | Claude 3 Haiku | Claude Opus 4.5 | Result |
|---|---|---|---|
| AIME26Source | — | 95.1% | Not comparable |
| HMMT Feb 2025Source | — | 92.9% | Not comparable |
| HMMT Nov 2025Source | — | 93.3% | Not comparable |
| HMMT Feb 2026Source | — | 85.3% | Not comparable |
| MMAnswerBenchSource | — | 84.0% | Not comparable |
| FrontierMath v2 (Tiers 1-3)Source | — | 20.690% | Not comparable |
| FrontierMath v2 (Tier 4)Source | — | 4.167% | Not comparable |
Multilingual2 benchmarks
Multimodal8 benchmarks
| Benchmark | Claude 3 Haiku | Claude Opus 4.5 | Result |
|---|---|---|---|
| AA-MMMU-ProSource | 30.8% | 71.2% | Claude Opus 4.5 leads |
| MMMU-ProSource | — | 70.6% | Not comparable |
| MathVisionSource | — | 74.3% | Not comparable |
| CharXivSource | — | 68.5% | Not comparable |
| VideoMMMUSource | — | 84.4% | Not comparable |
| ScreenSpot ProSource | — | 45.7% | Not comparable |
| V*Source | — | 67.0% | Not comparable |
| Design Arena WebsiteSource | — | 1277 | Not comparable |
Frequently Asked Questions (3)
Can I compare Claude 3 Haiku and Claude Opus 4.5 on BenchLM yet?
Not fully yet. BenchLM is tracking both models, but the sourced benchmark breakdown for this comparison is still coming soon.
Why does this comparison show “coming soon”?
BenchLM only shows category winners and benchmark-level calls when we have sourced results that can be compared fairly. For these models, the public benchmark coverage is not complete enough yet.
What data is available for Claude 3 Haiku and Claude Opus 4.5 today?
Claude 3 Haiku: $0.25 input / $1.25 output per 1M tokens Claude Opus 4.5: $5.00 input / $25.00 output per 1M tokens Both model pages still include creator, context window, reasoning mode, and other metadata while benchmark coverage fills in.
Related Comparisons
Choose a model with this week’s evidence
Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.
One email each week. Unsubscribe anytime.