Pairwise benchmark snapshot · Aug 1, 2026
In this head-to-head showdown, Gemini 2.5 Flash delivers higher overall intelligence and human-preferred responses (Elo 1,295 vs 1,224), while Gemini 2.5 Flash leads in SWE-bench software engineering benchmarks (48%), while Gemini 2.5 Flash is more budget-friendly at $0.4/1M per 1M output tokens. Review the complete breakdown below to determine which model best fits your performance and budget requirements.
8 direct benchmark disciplines evaluated across capability, speed, and cost
Previous cheap Claude. Haiku 4.5 is the current $1/$5 SKU.
Previous Google speed workhorse.
Δ 71 Arena Elo pts
185 tok/s
$0.4 / 1M output
Relative percentile scores computed across all active models in the benchmark catalog.
Comparing top Western standard models with China's leading frontier rival across 6 skill dimensions. Tap any spoke or dot to inspect.
Category wins across reasoning intelligence, generation speed, and token cost.
Preference Elo, Coding proficiency, SWE-bench & LiveBench accuracy
Generation throughput and time to first token responsiveness
Cost per million tokens and max context window length
| Benchmark | Claude 3.5 Haiku | Gemini 2.5 Flash | Advantage Delta |
|---|---|---|---|
| Preference Elo | +71 | ||
| LiveBench | +6.6% | ||
| SWE-bench | +11.8% | ||
| GPQA Diamond | +14.1% | ||
| Time to first token | — | ||
| Output speed | 140 tok/s | 185 tok/s | +45 tok/s |
| Input price | $0.8/1M | $0.1/1M | +$0.7/1M |
| Output price | $0.4/1M | +$3.6/1M | |
| Context window | +800k |
Simulate monthly production API costs in USD (US Dollar).
Gemini 2.5 Flash is estimated to save $78.50/month ($942/year).
Higher SWE-bench (48%).
Lower output list price ($0.4/1M).
Lower TTFT (110 ms).
Larger window (1M).
Both accept images. Defaulting to the higher-Elo side (Gemini 2.5 Flash).
| Target Workload | Recommended Pick | Evaluation Rationale |
|---|---|---|
| Repo / coding agents | Gemini 2.5 Flash | Higher SWE-bench (48%). |
| High-volume chat | Gemini 2.5 Flash | Lower output list price ($0.4/1M). |
| Voice / low-latency UI | Claude 3.5 Haiku | Lower TTFT (110 ms). |
| Long-document RAG | Gemini 2.5 Flash | Larger window (1M). |
| Screenshots / vision | Gemini 2.5 Flash | Both accept images. Defaulting to the higher-Elo side (Gemini 2.5 Flash). |
No comments posted on this matchup yet. Be the first to share an evaluation note!
Dynamic OpenGraph banner generated at /compare/claude-3-5-haiku-vs-gemini-2-5-flash/opengraph-image
Verified Head-to-Head Telemetry