Model comparison
Claude Haiku 4.5 vs Gemini 3.1 Pro
Head-to-head evidence from 3 shared benchmark results across 2 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: Claude Haiku 4.5 #77 (Estimated); Gemini 3.1 Pro #83 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. Claude Haiku 4.5 and Gemini 3.1 Pro share 3 comparable benchmark results. 1 of 8 categories are comparable. 2 results are unique to Claude Haiku 4.5; 44 to Gemini 3.1 Pro.
Updated July 20, 2026- Shared results
- 3
- Claude Haiku 4.5 only
- 2
- Gemini 3.1 Pro only
- 44
- Comparable categories
- 1 / 8
Pick Claude Haiku 4.5 if you want the stronger benchmark profile. Gemini 3.1 Pro only becomes the better choice if mathematics is the priority or you need the larger 1M context window.
Confidence note. This is a partial-evidence comparison with 3 shared benchmark results across 2 evidence categories; 1 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
Claude Haiku 4.5 has the cleaner BenchAlign overall profile here, landing at 56.58 versus 55.3. It is a real lead, but still close enough that category-level strengths matter more than the headline number.
Gemini 3.1 Pro is also the more expensive model on tokens at $2.00 input / $12.00 output per 1M tokens, versus $1.00 input / $5.00 output per 1M tokens for Claude Haiku 4.5. That is roughly 2.4x on output cost alone. Gemini 3.1 Pro gives you the larger context window at 1M, compared with 200K for Claude Haiku 4.5.
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 Haiku 4.5 | Δ | Gemini 3.1 Pro |
|---|---|---|---|
| Math | Claude Haiku 4.54.9 | Margin→ 26.9 | Gemini 3.1 Pro31.8 |
| Coding | Claude Haiku 4.573.3 | MarginNo overlap | Gemini 3.1 ProNot measured |
| Reasoning | Claude Haiku 4.5Not measured | MarginNo overlap | Gemini 3.1 Pro77.1 |
| Multimodal | Claude Haiku 4.5Not measured | MarginNo overlap | Gemini 3.1 Pro82.6 |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
FrontierMath v2 (Tiers 1-3)
MathA 5.903%B 36.900%Winner: Gemini 3.1 ProΔ 31FrontierMath v2 (Tiers 1-3): Claude Haiku 4.5 scored 5.903%; Gemini 3.1 Pro scored 36.900%. Gemini 3.1 Pro wins this benchmark. - Source ↗
FrontierMath v2 (Tier 4)
MathA 2.083%B 16.700%Winner: Gemini 3.1 ProΔ 14.6FrontierMath v2 (Tier 4): Claude Haiku 4.5 scored 2.083%; Gemini 3.1 Pro scored 16.700%. Gemini 3.1 Pro wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | Claude Haiku 4.5 | Gemini 3.1 Pro | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Claude Haiku 4.5$1 input / $5 output | Gemini 3.1 Pro$2 input / $12 output | Claude Haiku 4.5 has the lower combined listed price. |
| Generation speedtokens per second | Claude Haiku 4.5Not available | Gemini 3.1 Pro109 tok/s | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | Claude Haiku 4.5Not available | Gemini 3.1 Pro29.71 s | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | Claude Haiku 4.5200K | Gemini 3.1 Pro1M | Gemini 3.1 Pro lists the larger context window. |
Benchmark Deep Dive
Agentic17 benchmarks
| Benchmark | Claude Haiku 4.5 | Gemini 3.1 Pro | Result |
|---|---|---|---|
| JobBenchSource | 16.0% | — | Not comparable |
| Claw-EvalSource | — | 57.8% | Not comparable |
| DeepSearchQASource | — | 69.7% | Not comparable |
| τ²-bench resultsSource | — | 95.6% | Not comparable |
| AA Agentic IndexSource | — | 21.4% | Not comparable |
| APEX-Agents-AASource | — | 32.0% | Not comparable |
| GDPval-AASource | — | 23.2% | Not comparable |
| GDPval-AASource | — | 965 | Not comparable |
| Gert LabsSource | — | 56.87% | Not comparable |
| ResearchClawBenchSource | — | 13.3% | Not comparable |
| AA AutomationBenchSource | — | 37.5% | Not comparable |
| AA EnterpriseOps-GymSource | — | 42.2% | Not comparable |
| AA Harvey LABSource | — | 0.0% | Not comparable |
| AA ITBenchSource | — | 30.3% | Not comparable |
| AA Tau3 BankingSource | — | 16.5% | Not comparable |
| terminalBenchHardSource | — | 53.8% | Not comparable |
| aaTerminalBench21Source | — | 73.8% | Not comparable |
Coding6 benchmarks
| Benchmark | Claude Haiku 4.5 | Gemini 3.1 Pro | Result |
|---|---|---|---|
| SWE-bench VerifiedSource | 73.3% | — | Not comparable |
| LiveCodeBench ProSource | — | 82.9% | Not comparable |
| React Native EvalsSource | — | 78.9% | Not comparable |
| Vibe Code BenchSource | — | 32.03% | Not comparable |
| AA Coding IndexSource | — | 68.8% | Not comparable |
| AA-SciCodeSource | — | 58.9% | Not comparable |
Reasoning3 benchmarks
Knowledge10 benchmarks
| Benchmark | Claude Haiku 4.5 | Gemini 3.1 Pro | Result |
|---|---|---|---|
| GPQA-DSource | — | 94.3% | Not comparable |
| HLE w/o toolsSource | — | 45.4% | Not comparable |
| HealthBench HardSource | — | 20.6% | Not comparable |
| MedXpertQA (Text)Source | — | 71.5% | Not comparable |
| Artificial Analysis Intelligence IndexSource | — | 46.5% | Not comparable |
| AA-GPQA DiamondSource | — | 94.1% | Not comparable |
| AA-HLESource | — | 44.7% | Not comparable |
| AA-Omniscience IndexSource | — | 32.9% | Not comparable |
| AA-Omniscience AccuracySource | — | 55.3% | Not comparable |
| AA-Omniscience Hallucination RateSource | — | 49.9% | Not comparable |
MathGemini 3.1 Pro wins2 benchmarks
Multilingual1 benchmarks
| Benchmark | Claude Haiku 4.5 | Gemini 3.1 Pro | Result |
|---|---|---|---|
| AA Global-MMLU-LiteSource | — | 93.2% | Not comparable |
Multimodal9 benchmarks
| Benchmark | Claude Haiku 4.5 | Gemini 3.1 Pro | Result |
|---|---|---|---|
| Design Arena WebsiteSource | 1154 | 1284 | Gemini 3.1 Pro leads |
| MMMU-ProSource | — | 83.9% | Not comparable |
| CharXivSource | — | 80.2% | Not comparable |
| ERQASource | — | 69.4% | Not comparable |
| SimpleVQASource | — | 72.4% | Not comparable |
| ScreenSpot ProSource | — | 84.4% | Not comparable |
| ZeroBenchSource | — | 29.0% | Not comparable |
| MedXpertQA (MM)Source | — | 81.3% | Not comparable |
| AA-MMMU-ProSource | — | 82.4% | Not comparable |
Inst. Following1 benchmarks
| Benchmark | Claude Haiku 4.5 | Gemini 3.1 Pro | Result |
|---|---|---|---|
| AA-IFBenchSource | — | 77.1% | Not comparable |
Frequently Asked Questions (2)
Which is better, Claude Haiku 4.5 or Gemini 3.1 Pro?
Claude Haiku 4.5 is ahead on BenchLM's BenchAlign leaderboard, 56.58 to 55.3. The biggest single separator in this matchup is FrontierMath v2 (Tiers 1-3), where the scores are 5.903% and 36.900%.
Which is better for math, Claude Haiku 4.5 or Gemini 3.1 Pro?
Gemini 3.1 Pro has the edge for math in this comparison, averaging 31.8 versus 4.9. Inside this category, FrontierMath v2 (Tiers 1-3) is the benchmark that creates the most daylight between them.
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.