Model comparison
Claude Haiku 4.5 Thinking vs Llama 4 Maverick
Head-to-head evidence from 1 shared benchmark result across 1 category. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: Claude Haiku 4.5 Thinking unranked (Not scored); Llama 4 Maverick #191 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. Claude Haiku 4.5 Thinking and Llama 4 Maverick share 1 comparable benchmark result. 0 of 8 categories are comparable. 1 result is unique to Claude Haiku 4.5 Thinking; 17 to Llama 4 Maverick.
Updated July 23, 2026- Shared results
- 1
- Claude Haiku 4.5 Thinking only
- 1
- Llama 4 Maverick only
- 17
- Comparable categories
- 0 / 8
Benchmark data for Claude Haiku 4.5 Thinking and Llama 4 Maverick is coming soon on BenchLM.
Confidence note. This is a partial-evidence comparison with 1 shared benchmark result across 1 evidence category; 0 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
BenchLM has partial data for these models, but not enough overlapping benchmark coverage to produce a fair score-level comparison yet.
Llama 4 Maverick has the larger context window at 1M, compared with 200K for Claude Haiku 4.5 Thinking.
Category breakdown
Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.
| Category | Claude Haiku 4.5 Thinking | Δ | Llama 4 Maverick |
|---|---|---|---|
| Math | Claude Haiku 4.5 ThinkingNot measured | MarginNo overlap | Llama 4 Maverick0.7 |
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | Claude Haiku 4.5 Thinking | Llama 4 Maverick | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Claude Haiku 4.5 ThinkingNot available | Llama 4 Maverick$0 input / $0 output | A complete price comparison is not available. |
| Generation speedtokens per second | Claude Haiku 4.5 ThinkingNot available | Llama 4 Maverick121 tok/s | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | Claude Haiku 4.5 ThinkingNot available | Llama 4 Maverick0.95 s | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | Claude Haiku 4.5 Thinking200K | Llama 4 Maverick1M | Llama 4 Maverick lists the larger context window. |
Benchmark Deep Dive
Agentic4 benchmarks
Coding3 benchmarks
Reasoning2 benchmarks
Knowledge6 benchmarks
| Benchmark | Claude Haiku 4.5 Thinking | Llama 4 Maverick | Result |
|---|---|---|---|
| Artificial Analysis Intelligence IndexSource | — | 14.3% | Not comparable |
| AA-GPQA DiamondSource | — | 67.1% | Not comparable |
| AA-HLESource | — | 4.8% | Not comparable |
| AA-Omniscience IndexSource | — | -41.8% | Not comparable |
| AA-Omniscience AccuracySource | — | 24.3% | Not comparable |
| AA-Omniscience Hallucination RateSource | — | 87.3% | Not comparable |
Math1 benchmarks
| Benchmark | Claude Haiku 4.5 Thinking | Llama 4 Maverick | Result |
|---|---|---|---|
| FrontierMath v2 (Tiers 1-3)Source | — | 0.690% | Not comparable |
Multimodal2 benchmarks
Inst. Following1 benchmarks
| Benchmark | Claude Haiku 4.5 Thinking | Llama 4 Maverick | Result |
|---|---|---|---|
| AA-IFBenchSource | — | 43.0% | Not comparable |
Frequently Asked Questions (3)
Can I compare Claude Haiku 4.5 Thinking and Llama 4 Maverick on BenchLM yet?
Not fully yet. BenchLM is tracking both models, but the sourced benchmark breakdown for this comparison is still coming soon.
Why does this comparison show “coming soon”?
BenchLM only shows category winners and benchmark-level calls when we have sourced results that can be compared fairly. For these models, the public benchmark coverage is not complete enough yet.
What data is available for Claude Haiku 4.5 Thinking and Llama 4 Maverick today?
Llama 4 Maverick: $0.00 input / $0.00 output per 1M tokens Both model pages still include creator, context window, reasoning mode, and other metadata while benchmark coverage fills in.
Self-host vs API cost
Estimates at 50,000 req/day · 1000 tokens/req average.
Related Comparisons
Choose a model with this week’s evidence
Join 2,000+ readers for ranking moves, pricing changes, and the claims that still need proof.
One email each week. Unsubscribe anytime.