Claude Haiku 4.5 vs Gemini 3.5 Flash

Benchmarks, API pricing and specs, head to head. Data updated 2026-06-10.

Claude Haiku 4.5

Anthropic · Oct 2025

Anthropic's fast tier: Sonnet-4-class coding performance at $1/$5 and more than twice the speed.

Gemini 3.5 Flash

Google · May 2026

Google's I/O 2026 headliner: near-Pro intelligence at Flash speed, 78.8% on SWE-bench Verified with full multimodal input.

The verdict

Gemini 3.5 Flash wins 1 of the 1 benchmarks these models share, against 0 for Claude Haiku 4.5. Claude Haiku 4.5 is about 1.7x cheaper per blended million tokens (3:1 input:output mix). Gemini 3.5 Flash also takes 1M of context versus 200K for Claude Haiku 4.5.

Benchmark head-to-head 01

SWE-bench Verified
73.3%78.8%
Claude Haiku 4.5Gemini 3.5 Flash

Specs & pricing

Claude Haiku 4.5Gemini 3.5 Flash
modhub Index
Input price / 1M$1$1.5
Output price / 1M$5$9
Context window200K1M
Max output64K66K
Open weightsnono
Reasoning modelyesyes
Multimodal inputtext, imagetext, image, audio, video
Knowledge cutoffFeb 2025Feb 2026
ReleasedOct 2025May 2026
Example monthly cost*$17.50$28.50

* 10M input + 1.5M output tokens per month at list prices, no caching. Green = better value on that row.

Frequently asked questions

Which is better, Claude Haiku 4.5 or Gemini 3.5 Flash?
Gemini 3.5 Flash wins 1 of the 1 benchmarks these models share, against 0 for Claude Haiku 4.5. Claude Haiku 4.5 is about 1.7x cheaper per blended million tokens (3:1 input:output mix). Gemini 3.5 Flash also takes 1M of context versus 200K for Claude Haiku 4.5.
Which is cheaper, Claude Haiku 4.5 or Gemini 3.5 Flash?
Claude Haiku 4.5 costs $1/$5 per million input/output tokens, while Gemini 3.5 Flash costs $1.5/$9. For a typical workload of 10M input and 1.5M output tokens per month, that's $17.50 versus $28.50.
Which model is better for coding, Claude Haiku 4.5 or Gemini 3.5 Flash?
On SWE-bench Verified — the standard agentic-coding benchmark — Gemini 3.5 Flash scores 78.8% versus 73.3% for Claude Haiku 4.5, making Gemini 3.5 Flash the stronger pick for coding agents.

More comparisons