Granite 4 H Medium vs Mistral Large 3
Side-by-side comparison of API pricing, specs, benchmarks, and capabilities
| Specification |
by IBM
|
by Mistral
|
|---|---|---|
| Overview | ||
| Status | Current mid tier Open weights | Current mid tier |
| Released | Oct 19, 2025 | Nov 1, 2025 |
| Pricing per million tokens | ||
| Input | $0.150/Mtok | $0.500/Mtok |
| Output | $0.600/Mtok | $1.50/Mtok |
| Blended avg | $0.262/Mtok | $0.750/Mtok |
| Specifications | ||
| Context window | 128K tokens | 128K tokens |
| Parameters | Proprietary | Proprietary |
| Speed (TPS) | — | — |
| Modalities | ||
| Input |
text
|
text
|
| Benchmarks sources: IBM model card / Mistral model card, Artificial Analysis | ||
| GPQA Diamond | 35 vendor | 58 |
| SWE-Bench Verified | 15 vendor | 35 |
| Humanity's Last Exam | 7 vendor | 18 |
| ARC-AGI 2 | 5 vendor | 20 |
| AIME 2025 | 28 vendor | 52 |
| MMMLU | 68 vendor | 78 |
| BFCL | 48 vendor | 68 |
| HumanEval | 72 vendor | 85 |
| MATH 500 | 50 vendor | 68 |
| Avg benchmark score | — | 53.6 |
| Perf / dollar | — | 71.5 |
| Independent composite scores each on its own scale — not part of the average above | ||
| AA Intelligence Index | — | 15.9 |
| AA Agentic Index | — | 5.5 |
| AA-Omniscience Index (−100–100) | — | -39.6 |
| LMSYS Chatbot Arena | — | 1415.3 ELO |
| Providers | ||
| Available from |
IBM — $0.150/$0.600/Mtok
|
Mistral — $0.500/$1.50/Mtok
|
Cost at scale
1M tokens · 50/50 input/output| Volume | Granite 4 H Medium | Mistral Large 3 | Savings |
|---|---|---|---|
| 1M tokens | $0.26 | $0.75 | $0.49 (65.3%) |
| 10M tokens | $2.62 | $7.5 | $4.88 (65.1%) |
| 100M tokens | $26.25 | $75 | $48.75 (65%) |
| 1000M tokens | $262.5 | $750 | $487.5 (65%) |
When to pick which
distilled from the pricing and spec data above- cost dominates: $0.262/Mtok blended vs $0.750 — 65% less on the same 50/50 token mix
- you want open weights — self-host it, fine-tune it, or exit the API entirely; Mistral Large 3 is closed
- this pairing gives Mistral Large 3 no edge on price, context, cache discount, or measured performance — pick it only for qualitative fit (ecosystem, compliance, model behavior on your prompts)
The cheaper option here — Granite 4 H Medium costs $0.262/Mtok blended on IBM.
Summary
Granite 4 H Medium by IBM costs $0.150/Mtok input and $0.600/Mtok output, with a 128K-token context window. It supports text input.
Mistral Large 3 by Mistral costs $0.500/Mtok input and $1.50/Mtok output, with a 128K-token context window. It supports text input.
On a blended cost basis, Granite 4 H Medium is 65% cheaper than Mistral Large 3.
The two aren't directly comparable on average benchmark score: Mistral Large 3 has published per-benchmark results, while Granite 4 H Medium does not yet.
Note: Pricing is per million tokens. Actual costs vary with usage patterns, prompt caching, and batch discounts. Always verify against official provider pricing pages.