Model comparison
Claude Haiku 4.5 vs Mistral Large
Claude Haiku 4.5 is the stronger model overall, scoring 39.5 to 31.9 on the Noometry Index.
Last verified . 35 shared benchmarks.
Mistral AISummary
- They share 35 benchmarks with published results for both. Claude Haiku 4.5 scores higher in 8 categories and Mistral Large in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Haiku 4.5 leads 44.9 to 18.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 66.7% for Claude Haiku 4.5 and 8.5% for Mistral Large.
- Claude Haiku 4.5 is cheaper at $1 / $5 per million input/output tokens, against $2 / $6 for Mistral Large.
- Claude Haiku 4.5 accepts more context: 200K tokens versus 131K.
- Mistral Large has downloadable open weights; the other is API-only.
Side by side
| Claude Haiku 4.5 | Mistral Large | |
|---|---|---|
| Provider | Anthropic | Mistral AI |
| Noometry Index | 39.5 | 31.9 |
| Released | 2025-10-15 | 2024-02-26 |
| Weights | Proprietary | Open |
| Context window | 200K | 131K |
| Max output | 64K | 16K |
| Input $ / M tokens | $1 | $2 |
| Output $ / M tokens | $5 | $6 |
| Results tracked | 53 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Haiku 4.5 leads
Claude Haiku 4.5: 44.0 (#78), Mistral Large: 34.3 (#240)
| Benchmark | Claude Haiku 4.5 | Mistral Large |
|---|---|---|
| SciCode | 43.3% | 36.2% |
| LMArena Coding | 1453 | 1277 |
| ALE-Bench | 653.48 | 264.7 |
| SWE-bench Verified (bash only) | 66.6% | — |
| LMArena WebDev | 1330 | — |
| SWE-bench Multilingual | 64.7% | — |
| WeirdML | 45.4% | — |
| BigCodeBench Instruct | — | 30% |
| LiveBench Coding | — | 47.1% |
| BigCodeBench Complete | — | 38.3% |
| HumanEval+ | — | 62.2% |
| MBPP+ | — | 59.5% |
Agentic & Tool Use Claude Haiku 4.5 leads
Claude Haiku 4.5: 33.6 (#52), Mistral Large: 28.6 (#89)
| Benchmark | Claude Haiku 4.5 | Mistral Large |
|---|---|---|
| Berkeley Function Calling Leaderboard | 68.7% | 38.4% |
| Terminal-Bench | 35.5% | — |
| DeepResearch Bench | 45.5% | — |
| BALROG | 31.2% | — |
| ExploitBench | 13.7% | — |
| Vending-Bench 2 | 458.89 | — |
Reasoning Too close to call
Claude Haiku 4.5: 15.1 (#320), Mistral Large: 15.8 (#310)
| Benchmark | Claude Haiku 4.5 | Mistral Large |
|---|---|---|
| CritPt | 0% | 0% |
| LMArena Hard Prompts | 1420 | 1257 |
| DTBench | 73.6% | 65.1% |
| LMCA | 30.9% | 16.7% |
| Epoch Capabilities Index | 142.41 | 128.52 |
| ForecastBench | 61.4 | 57.1 |
| ARC-AGI-2 | 4% | — |
| SimpleBench | — | 22.5% |
| NYT Connections (extended) | 14.3% | — |
| ARC-AGI-1 | 47.7% | — |
| Chess Puzzles | 8% | — |
| LiveBench Reasoning | — | 43.5% |
| LiveBench Data Analysis | — | 50.1% |
| LiveBench | — | 48.4% |
Math Claude Haiku 4.5 leads
Claude Haiku 4.5: 44.9 (#78), Mistral Large: 18.2 (#291)
| Benchmark | Claude Haiku 4.5 | Mistral Large |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.7% | 8.5% |
| Omni-MATH | 56.1% | 28.1% |
| LMArena Math | 1396 | 1262 |
| MATH Level 5 | 96.4% | 50.3% |
| FrontierMath (Feb 2025 set) | 5.9% | 0.3% |
| LiveBench Math | — | 42.5% |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Claude Haiku 4.5 leads
Claude Haiku 4.5: 37.7 (#153), Mistral Large: 30.1 (#230)
| Benchmark | Claude Haiku 4.5 | Mistral Large |
|---|---|---|
| GPQA Diamond | 71.2% | 51.3% |
| MMLU-Pro | 77.7% | 59.9% |
| Vectara Hallucination Rate | 9.8% | 4.5% |
| GPQA (HELM) | 60.5% | 43.5% |
| LMArena Expert | 1442 | 1232 |
| SimpleQA Verified | 13.2% | — |
| Confabulations | — | 21.4% |
| MMLU | — | 80% |
Multimodal Not comparable
Claude Haiku 4.5: 26.8 (#118), Mistral Large: —
| Benchmark | Claude Haiku 4.5 | Mistral Large |
|---|---|---|
| Blueprint-Bench 2 | 0% | — |
| LMArena Document | 1420 | — |
Multilingual Claude Haiku 4.5 leads
Claude Haiku 4.5: 49.9 (#129), Mistral Large: 40.0 (#219)
| Benchmark | Claude Haiku 4.5 | Mistral Large |
|---|---|---|
| LMArena Non-English | 1377 | 1237 |
| LMArena Chinese | 1417 | 1240 |
| LMArena French | 1408 | 1325 |
| LMArena German | 1375 | 1254 |
| LMArena Japanese | 1339 | 1188 |
| LMArena Korean | 1347 | 1202 |
| LMArena Russian | 1381 | 1257 |
| LMArena Spanish | 1420 | 1268 |
Instruction Following Claude Haiku 4.5 leads
Claude Haiku 4.5: 71.4 (#149), Mistral Large: 67.9 (#191)
| Benchmark | Claude Haiku 4.5 | Mistral Large |
|---|---|---|
| IFEval | 80.1% | 87.7% |
| LMArena Instruction Following | 1414 | 1249 |
| LiveBench Instruction Following | — | 67.9% |
Long Context Claude Haiku 4.5 leads
Claude Haiku 4.5: 43.6 (#92), Mistral Large: 38.3 (#199)
| Benchmark | Claude Haiku 4.5 | Mistral Large |
|---|---|---|
| LMArena Longer Query | 1427 | 1261 |
Writing & Preference Claude Haiku 4.5 leads
Claude Haiku 4.5: 57.9 (#123), Mistral Large: 40.7 (#242)
| Benchmark | Claude Haiku 4.5 | Mistral Large |
|---|---|---|
| LMArena Text | 1396 | 1266 |
| LMArena Creative Writing | 1372 | 1243 |
| WildBench | 83.9% | 80.1% |
| LMArena Multi-Turn | 1409 | 1260 |
| Short-Story Creative Writing | — | 69% |
| EQ-Bench Creative Writing | — | 985 |
| EQ-Bench 4 | 1064 | — |
| LiveBench Language | — | 39.4% |
Frequently asked questions
Is Claude Haiku 4.5 better than Mistral Large?
Claude Haiku 4.5 is the stronger model overall, scoring 39.5 to 31.9 on the Noometry Index.
Which is cheaper, Claude Haiku 4.5 or Mistral Large?
Claude Haiku 4.5 is cheaper. It lists at $1 per million input tokens and $5 per million output tokens; Mistral Large lists at $2 and $6.
Is Claude Haiku 4.5 or Mistral Large better for coding?
Claude Haiku 4.5 scores higher on coding benchmarks: 44.0 versus 34.3 in the Noometry coding category.
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 Mistral Large share?
35 benchmarks have published results for both models. Claude Haiku 4.5 has 53 scored results on Noometry and Mistral Large has 51.