Model comparison
Claude Haiku 4.5 vs Granite 4.0 Micro
Claude Haiku 4.5 is the stronger model overall, scoring 39.3 to 28.8 on the Noometry Index. Granite 4.0 Micro costs 49× less per token, which makes it the better buy when Claude Haiku 4.5's lead doesn't matter for your workload.
Last verified . 8 shared benchmarks.
Summary
- They share 8 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 4 categories and Granite 4.0 Micro in 1 category; 5 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Haiku 4.5 leads 43.0 to 11.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 66.7% for Claude Haiku 4.5 and 2.8% for Granite 4.0 Micro.
- Granite 4.0 Micro is cheaper at $0.017 / $0.11 per million input/output tokens, against $1 / $5 for Claude Haiku 4.5.
- Claude Haiku 4.5 accepts more context: 200K tokens versus 131K.
- Granite 4.0 Micro has downloadable open weights; the other is API-only.
Side by side
| Claude Haiku 4.5 | Granite 4.0 Micro | |
|---|---|---|
| Provider | Anthropic | IBM |
| Noometry Index | 39.3 | 28.8 |
| Released | 2025-10-15 | 2025-10-02 |
| Weights | Proprietary | Open |
| Context window | 200K | 131K |
| Max output | 64K | 118K |
| Input $ / M tokens | $1 | $0.017 |
| Output $ / M tokens | $5 | $0.11 |
| Results tracked | 53 | 8 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Claude Haiku 4.5: 43.9 (#71), Granite 4.0 Micro: —
| Benchmark | Claude Haiku 4.5 | Granite 4.0 Micro |
|---|---|---|
| SWE-bench Verified (bash only) | 66.6% | — |
| LMArena WebDev | 1330 | — |
| SWE-bench Multilingual | 64.7% | — |
| SciCode | 43.3% | — |
| WeirdML | 45.4% | — |
| LMArena Coding | 1454 | — |
| ALE-Bench | 653.48 | — |
Agentic & Tool Use Not comparable
Claude Haiku 4.5: 33.5 (#50), Granite 4.0 Micro: —
| Benchmark | Claude Haiku 4.5 | Granite 4.0 Micro |
|---|---|---|
| Terminal-Bench | 35.5% | — |
| Berkeley Function Calling Leaderboard | 68.7% | — |
| DeepResearch Bench | 45.5% | — |
| BALROG | 31.2% | — |
| ExploitBench | 13.7% | — |
| Vending-Bench 2 | 458.89 | — |
Reasoning Granite 4.0 Micro leads
Claude Haiku 4.5: 15.1 (#310), Granite 4.0 Micro: 18.5 (#261)
| Benchmark | Claude Haiku 4.5 | Granite 4.0 Micro |
|---|---|---|
| Chess Puzzles | 8% | 0% |
| ARC-AGI-2 | 4% | — |
| NYT Connections (extended) | 14.3% | — |
| ARC-AGI-1 | 47.7% | — |
| CritPt | 0% | — |
| LMArena Hard Prompts | 1420 | — |
| DTBench | 73.6% | — |
| LMCA | 30.9% | — |
| Epoch Capabilities Index | 142.42 | — |
| ForecastBench | 61.4 | — |
Math Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.0 (#69), Granite 4.0 Micro: 11.2 (#301)
| Benchmark | Claude Haiku 4.5 | Granite 4.0 Micro |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.7% | 2.8% |
| Omni-MATH | 56.1% | 20.9% |
| LMArena Math | 1395 | — |
| MATH Level 5 | 96.4% | — |
| FrontierMath (Feb 2025 set) | 5.9% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Claude Haiku 4.5 leads
Claude Haiku 4.5: 37.6 (#146), Granite 4.0 Micro: 9.9 (#299)
| Benchmark | Claude Haiku 4.5 | Granite 4.0 Micro |
|---|---|---|
| GPQA Diamond | 71.2% | 28.3% |
| MMLU-Pro | 77.7% | 39.5% |
| GPQA (HELM) | 60.5% | 30.7% |
| SimpleQA Verified | 13.2% | — |
| Vectara Hallucination Rate | 9.8% | — |
| LMArena Expert | 1443 | — |
Multimodal Not comparable
Claude Haiku 4.5: 27.0 (#111), Granite 4.0 Micro: —
| Benchmark | Claude Haiku 4.5 | Granite 4.0 Micro |
|---|---|---|
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1420 | — |
Multilingual Not comparable
Claude Haiku 4.5: 50.1 (#124), Granite 4.0 Micro: —
| Benchmark | Claude Haiku 4.5 | Granite 4.0 Micro |
|---|---|---|
| LMArena Non-English | 1376 | — |
| LMArena Chinese | 1415 | — |
| LMArena French | 1408 | — |
| LMArena German | 1377 | — |
| LMArena Japanese | 1339 | — |
| LMArena Korean | 1346 | — |
| LMArena Russian | 1380 | — |
| LMArena Spanish | 1420 | — |
Instruction Following Claude Haiku 4.5 leads
Claude Haiku 4.5: 71.6 (#144), Granite 4.0 Micro: 70.0 (#164)
| Benchmark | Claude Haiku 4.5 | Granite 4.0 Micro |
|---|---|---|
| IFEval | 80.1% | 84.9% |
| LMArena Instruction Following | 1414 | — |
Long Context Not comparable
Claude Haiku 4.5: 43.6 (#86), Granite 4.0 Micro: —
| Benchmark | Claude Haiku 4.5 | Granite 4.0 Micro |
|---|---|---|
| LMArena Longer Query | 1427 | — |
Writing & Preference Claude Haiku 4.5 leads
Claude Haiku 4.5: 58.1 (#119), Granite 4.0 Micro: 46.7 (#212)
| Benchmark | Claude Haiku 4.5 | Granite 4.0 Micro |
|---|---|---|
| WildBench | 83.9% | 67% |
| LMArena Text | 1396 | — |
| LMArena Creative Writing | 1371 | — |
| EQ-Bench 4 | 1064 | — |
| LMArena Multi-Turn | 1409 | — |
Frequently asked questions
Is Claude Haiku 4.5 better than Granite 4.0 Micro?
Claude Haiku 4.5 is the stronger model overall, scoring 39.3 to 28.8 on the Noometry Index. Granite 4.0 Micro costs 49× less per token, which makes it the better buy when Claude Haiku 4.5's lead doesn't matter for your workload.
Which is cheaper, Claude Haiku 4.5 or Granite 4.0 Micro?
Granite 4.0 Micro is cheaper. It lists at $0.017 per million input tokens and $0.11 per million output tokens; Claude Haiku 4.5 lists at $1 and $5.
Which has the bigger context window?
Claude Haiku 4.5 does, with 200K tokens against 131K.
How many benchmarks do Claude Haiku 4.5 and Granite 4.0 Micro share?
8 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and Granite 4.0 Micro has 8.