Loading benchmark catalogue
Compare performance benchmarks, coding intelligence, reasoning evaluations, context limits, and API economics between Gemini 3.8 Flash (Google DeepMind) and Claude Haiku 5.5 (Anthropic).
| Metric / Capability | Gemini 3.8 Flash | Claude Haiku 5.5 |
|---|---|---|
| Developer / Organization | Google DeepMind | Anthropic |
| Release Date | 2026-09-02 | 2026-10-07 |
| Deployment / Weights | Proprietary Cloud API | Proprietary Cloud API |
| Context Window | 1.0M tokens (1,048,576 tokens) | 1M tokens (1,000,000 tokens) |
| Input Pricing (/1M Tokens) | $0.825 | $0.1 |
| Output Pricing (/1M Tokens) | $4.125 | $0.5 |
| Benchmark / Domain | Gemini 3.8 Flash | Claude Haiku 5.5 | Advantage |
|---|---|---|---|
Artificial Analysis Intelligence Index Composite index aggregating 10+ evaluations of general reasoning and knowledge. | 40.9% | 43.4% | Claude Haiku 5.5 (+6%) |
Artificial Analysis Coding Index Composite index of coding and software-engineering evaluations. | 76.3% | — | — |
Artificial Analysis Agentic Index Composite index of tool-use and long-horizon agentic tasks. | 40.2% | — | — |
Artificial Analysis Coding Agent Index v2.0 Composite index of coding-agent systems (model + harness) across FrontierCode v1.1, SWE-Atlas-QnA, and Terminal-Bench v4. | 33.3% | 37.3% | Claude Haiku 5.5 (+12%) |
Terminal-Bench 4.0 Sandboxed command-line tasks requiring shell and file-system operation. | 19.1% | 39.2% | Claude Haiku 5.5 (+105%) |
ARC-AGI-2 Novel visual grid puzzles requiring abstract rule induction. | 89.2% | 61.3% | Gemini 3.8 Flash (+46%) |
Humanity's Last Exam Expert-written questions across hundreds of disciplines designed to resist saturation. | — | 45.9% | — |
FrontierMath v2 (Tiers 1-3) Unpublished, research-level mathematics problems across Tiers 1-3. | 68.4% | 75.1% | Claude Haiku 5.5 (+10%) |
Design Arena ELO Human-preference ELO across UI, 3D, data-visualization, and game-dev generations. | 1313 | 1344 | Claude Haiku 5.5 (+2%) |
Claude Haiku 5.5 outperforms Gemini 3.8 Flash across 5 evaluated benchmarks compared to 1 for Gemini 3.8 Flash, excelling in Artificial Analysis Intelligence Index, Artificial Analysis Coding Agent Index v2.0, Terminal-Bench 4.0.
Gemini 3.8 Flash offers a larger context capacity of 1.0M tokens (1,048,576 tokens), compared to 1M tokens for Claude Haiku 5.5.
Claude Haiku 5.5 is more economical. Input tokens cost $0.1/M tokens compared to $0.825/M tokens for Gemini 3.8 Flash.
Gemini 3.8 Flash is a proprietary cloud API. Claude Haiku 5.5 is a proprietary cloud API.
Analyze 130+ models side-by-side using interactive radar charts, cost-efficiency quadrant plots, and live dataset filters on the full NeuBench index.
Open Interactive Comparator