Claude Haiku 5.5 vs Gemini 4 Carbon
Gemini 4 Carbon leads Claude Haiku 5.5 by 16.1 points on the Shogo Agent Index. Claude Haiku 5.5 wins 0 and Gemini 4 Carbon wins 15 of 15 shared benchmarks. Claude Haiku 5.5 is cheaper for agent workloads.
Add more models to this comparison →Pillars
| Pillar | Claude Haiku 5.5 | Gemini 4 Carbon |
|---|---|---|
| Agentic coding | 67 | 82 |
| Long-horizon autonomy | 62 | 80 |
| Knowledge-work agents | 46 | 63 |
| Computer use | 47 | 68 |
| Security & context learning | 55 | 63 |
Shared benchmarks
| Benchmark | Claude Haiku 5.5 | Gemini 4 Carbon |
|---|---|---|
| APEX-Agentsindex | 56.1% | 84.2% |
| Vending-Bench 2index | $3,602 | $16,484 |
| FrontierSWEindex | 43.8% | 61.2% |
| CursorBenchindex | 48.4% | 60.4% |
| ARC-AGI | 89.0% | 98.5% |
| ARC-AGI-2 | 61.3% | 94.2% |
| Chess Puzzles | 55.0% | 78.0% |
| CritPt | 22.6% | 36.0% |
| FrontierMath Tier 4 | 46.3% | 95.1% |
| FrontierMath Tiers 1-3 | 75.1% | 90.5% |
| GPQA Diamond | 89.6% | 96.5% |
| OTIS Mock AIME | 98.9% | 100.0% |
| SciCode | 55.0% | 64.2% |
| SimpleBench | 59.2% | 92.6% |
| SimpleQA Verified | 23.8% | 77.7% |
Specs and price
| Claude Haiku 5.5 | Gemini 4 Carbon | |
|---|---|---|
| Agent Index | 57.5 (#18) | 73.6 (#2) |
| Agent $/M | $0.050 | $0.84 |
| Input / output $/M | $0.10 / $0.50 | $2.00 / $10.00 |
| Context | 1M | 1M |
| Released | Oct 7, 2026 | Nov 10, 2026 |
| Weights | Closed | Closed |
| API hosts | 5 | 5 |
Don’t pick one. Route.
Shogo can run Claude Haiku 5.5 and Gemini 4 Carbon side by side and send each step of an agent to the one that’s cheaper for it.
Try both in Shogo