Claude 3.5 HaikuvsGrok‑1
Claude 3.5 Haiku | Grok‑1 | |
|---|---|---|
| BenchmarksPublished by one model only | ||
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. | 50% | — |
SWE-Bench VerifiedCoding — Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. | 40.6% | — |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 41.6% | — |
| Overview | ||
| Company | Anthropic | SpaceXAI |
| Release date | Oct 22 2024 | Nov 3 2023 |
| Access | Proprietary | Open Weight |
Other comparisons
Claude 3.5 HaikuvsGPT-6 AstraGrok‑1vsGPT-6 AstraClaude 3.5 HaikuvsGemini 3.8 FlashGrok‑1vsGemini 3.8 FlashClaude 3.5 HaikuvsMuse Spark 1.3Grok‑1vsMuse Spark 1.3Claude 3.5 HaikuvsDeepSeek-V4.1-FlashGrok‑1vsDeepSeek-V4.1-FlashClaude 3.5 HaikuvsMistral Medium 3.5Grok‑1vsMistral Medium 3.5Claude 3.5 HaikuvsKimi K3Grok‑1vsKimi K3
Frequently asked questions
Claude 3.5 Haiku and Grok‑1 don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. Grok‑1 shipped 354 days before Claude 3.5 Haiku, so benchmark comparisons should account for the intervening progress.
Claude 3.5 Haiku is proprietary, while Grok‑1 is open weight.
Direct benchmark comparisons are unavailable — Claude 3.5 Haiku and Grok‑1 don't publish scores on any of the same benchmarks.