Gemini 3.5 Flash-Lite vs Claude Haiku 4.5
Side-by-side comparison of API pricing, specs, benchmarks, and capabilities
| Specification |
by Google
|
by Anthropic
|
|---|---|---|
| Overview | ||
| Status | Current budget | Current budget |
| Released | ||
| Pricing per million tokens | ||
| Input | $0.300/Mtok | $1.00/Mtok |
| Output | $2.50/Mtok | $5.00/Mtok |
| Blended avg | $0.850/Mtok | $2.00/Mtok |
| Cached input | $0.030/Mtok | $0.100/Mtok |
| Price basis | Verified against Google's Gemini API pricing page (ai.google.dev) on 2026-07-28: the Gemini 3.5 Flash-Lite Standard paid tier is $0.30/1M input, $2.50/1M output, $0.03/1M context caching (batch tier $0.15/$1.25). The identical $0.30/$2.50 price shared with the deprecated Gemini 2.5 Flash reflects Google's uniform Flash tier pricing, not a placeholder — the value matches LiteLLM gemini/gemini-3.5-flash-lite. | — |
| Specifications | ||
| Context window | 1M tokens | 200K tokens |
| Parameters | Proprietary | Proprietary |
| Speed (TPS) | — | — |
| Modalities | ||
| Input |
textimageaudiovideo
|
textimage
|
| Benchmarks sources: Artificial Analysis, Google DeepMind Gemini 3.5 Flash-Lite model card, LMSYS Chatbot Arena (UC Berkeley) / Vellum LLM Leaderboard | ||
| Terminal-Bench 2.1 | 54 vendor | 27.5 |
| Avg benchmark score | — | 65.5 |
| Perf / dollar | — | 32.8 |
| GPQA Diamond | — | 73 |
| MCP Atlas | — | 40.2 |
| OSWorld-Verified | 74 vendor | — |
| AIME 2025 | — | 96.3 |
| MMMLU | — | 83 |
| HumanEval | — | 73.3 |
| MATH 500 | — | 78 vendor |
| Independent composite scores each on its own scale — not part of the average above | ||
| AA Intelligence Index | 37.4 | 29.9 |
| AA Agentic Index | 27.2 | 16.5 |
| AA-Omniscience Index (−100–100) | 5.2 | -4.4 |
| LMSYS Chatbot Arena | 1456.9 ELO | 1413.2 ELO |
| GDPval-AA v2 | 1136.3 ELO | — |
| Providers | ||
| Available from |
Google — $0.300/$2.50/Mtok
|
Anthropic — $1.00/$5.00/Mtok
|
Cost at scale
1M tokens · 50/50 input/output| Volume | Gemini 3.5 Flash-Lite | Claude Haiku 4.5 | Savings |
|---|---|---|---|
| 1M tokens | $0.85 | $2 | $1.15 (57.5%) |
| 10M tokens | $8.5 | $20 | $11.5 (57.5%) |
| 100M tokens | $85 | $200 | $115 (57.5%) |
| 1000M tokens | $850 | $2000 | $1150 (57.5%) |
When to pick which
distilled from the pricing and spec data above- cost dominates: $0.850/Mtok blended vs $2.00 — 57.5% less on the same 50/50 token mix
- you need the longer context: 1M tokens vs 200K (5×)
- this pairing gives Claude Haiku 4.5 no edge on price, context, cache discount, or measured performance — pick it only for qualitative fit (ecosystem, compliance, model behavior on your prompts)
The cheaper option here — Gemini 3.5 Flash-Lite costs $0.850/Mtok blended on Google.
Summary
Gemini 3.5 Flash-Lite by Google costs $0.300/Mtok input and $2.50/Mtok output, with a 1M-token context window. It supports text, image, audio, video input.
Claude Haiku 4.5 by Anthropic costs $1.00/Mtok input and $5.00/Mtok output, with a 200K-token context window. It supports text, image input.
On a blended cost basis, Gemini 3.5 Flash-Lite is 57.5% cheaper than Claude Haiku 4.5.It also has a larger context window.
The two aren't directly comparable on average benchmark score: Claude Haiku 4.5 has published per-benchmark results, while Gemini 3.5 Flash-Lite does not yet — it is measured today on independent composites (see the table above).
Note: Pricing is per million tokens. Actual costs vary with usage patterns, prompt caching, and batch discounts. Always verify against official provider pricing pages.