Claude 3 HaikuvsDeepSeek R1-0528
Claude 3 Haiku | DeepSeek R1-0528 | |
|---|---|---|
| API pricingUSD per 1M tokens · lower wins | ||
Cheapest inputLowest input rate across third-party providers, excluding the lab itself. The cheapest endpoint may run a quantised build or a shorter context — see "Available from" on the model page. | $0.25Amazon Bedrock | $0.50DeepInfra |
Cheapest outputLowest output rate across third-party providers, excluding the lab itself. May come from a different provider than the cheapest input. | $1.25Amazon Bedrock | $2.15DeepInfra |
| Benchmarks | ||
BullshitBench v2Nonsense detection — Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better. | 10% | 8% |
| BenchmarksPublished by one model only | ||
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 33.3% | — |
| Overview | ||
| Company | Anthropic | DeepSeek |
| Release date | Mar 4 2024 | May 28 2025 |
| Access | Proprietary | Open Weight |
Other comparisons
Claude 3 HaikuvsGPT-6 AstraDeepSeek R1-0528vsGPT-6 AstraClaude 3 HaikuvsGemini 3.8 FlashDeepSeek R1-0528vsGemini 3.8 FlashClaude 3 HaikuvsMuse Spark 1.3DeepSeek R1-0528vsMuse Spark 1.3Claude 3 HaikuvsGrok 4.6DeepSeek R1-0528vsGrok 4.6Claude 3 HaikuvsMistral Medium 3.5DeepSeek R1-0528vsMistral Medium 3.5Claude 3 HaikuvsKimi K3DeepSeek R1-0528vsKimi K3
Frequently asked questions
Claude 3 Haiku leads DeepSeek R1-0528 on 1 of the 1 benchmark they both report (BullshitBench v2). Claude 3 Haiku shipped 450 days before DeepSeek R1-0528, so benchmark comparisons should account for the intervening progress.
Claude 3 Haiku is proprietary, while DeepSeek R1-0528 is open weight.
On BullshitBench v2, Claude 3 Haiku leads at 10% vs DeepSeek R1-0528 at 8%.