Compare

Claude Haiku 4.5 vs Mistral X-Large

Anthropic vs Mistral. Specs, benchmarks, and real per-task cost — all in one page.

Verdict

Mistral X-Large leads on LMSYS ELO (1318 vs 1305). Claude Haiku 4.5 is ~1.9× cheaper on a 3:1 input:output blend. Mistral X-Large is open-weights, the other is proprietary.

Anthropic

Claude Haiku 4.5

Fast, cheap, and smart enough for most routing and extraction tasks. Great base model for subagents.

Sub-second latency10x cheaper than OpusStrong structured output

Mistral

Mistral X-Large

European frontier model. EU-hosted inference, strong European language coverage, Apache-licensed weights.

EU data residencyMultilingual strengthOpen license

Pricing

Input / 1M$0.80$2.00

Output / 1M$4.00$6.00

Context200K256K

Max output16K16K

LicenseProprietaryOpen weights

Released2025-10-012026-01-28

Benchmarks

LMSYS ELO1305.01318.0

MMLU Pro81.483.5

HumanEval87.088.0

SWE-Bench49.2—

MATH76.878.2

IFEval87.386.5

Per-task cost

Summarize a 1-hour meeting transcript$0.014$0.033

Review a 500-line pull request$0.014$0.028

Answer a customer support ticket$0.0056$0.011

Extract structured data from a resume$0.0080$0.016

Debug a stack trace with context$0.013$0.024

Per-call cost using published token counts for each task. Real-world prompts vary.

Tool

Interactive compare →

Add up to four models, tweak tasks live.