kimi-k2-0905-preview vs mistral-large-3
mistral-large-3 is 1.2× cheaper on input than kimi-k2-0905-preview.
| kimi-k2-0905-preview | mistral-large-3 | |
|---|---|---|
| Input / 1M | $0.60 | $0.50 |
| Cached input / 1M | $0.15 | — |
| Output / 1M | $2.50 | $1.50 |
| Context window | 262K | 262K |
| Max output | 262K | 262K |
| Provider | Moonshot | Mistral |
| Vision | No | Yes |
| Tool calling | Yes | Yes |
| Token count | estimated | estimated |
Capability for the money
Independent measurement · Artificial Analysis · read 2026-08-11Not the same tier: kimi-k2-0905-preview leads mistral-large-3 by 48.8 points on tool use (τ²-bench), for 1.4× the price.
Following instructions
extraction, classification, sticking to a format
IFBenchkimi-k2-0905-preview +5.5
Coding
writing and fixing code
LiveCodeBenchkimi-k2-0905-preview +14.5
Tool use
multi-step agent work against real tools
τ²-benchkimi-k2-0905-preview +48.8
Hard reasoning
graduate-level science and multi-step analysis
GPQA Diamondkimi-k2-0905-preview +8.7
Averaged over the benchmarks both models sat, kimi-k2-0905-preview returns 58.8 points of pass rate per dollar of blended price against 58.4 — the ratio that decides whether the dearer model earns its rate.
Each figure is a pass rate: the share of that test set the model answered correctly, measured by a third party rather than self-reported by the lab. Rows are different tests of different difficulty, so compare a model against the other model, not one row against the next. And a benchmark is not your workload — it narrows the shortlist, it does not pick for you.
Which to pick
- Cheaper input: mistral-large-3
- Cheaper output: mistral-large-3
- Larger context: kimi-k2-0905-preview
- Better at following instructions: kimi-k2-0905-preview (41.7% vs 36.2% on IFBench)
- Better at coding: kimi-k2-0905-preview (61.0% vs 46.5% on LiveCodeBench)
- Better at tool use: kimi-k2-0905-preview (73.4% vs 24.6% on τ²-bench)
- Better at hard reasoning: kimi-k2-0905-preview (76.7% vs 68.0% on GPQA Diamond)
List price is not production cost. Output length, cache hit rate, retries and task quality move the real number more than the headline rate does — price your own prompt on the token ledger before deciding.
Related comparisons
kimi-k2-0905-preview detail · mistral-large-3 detail · All Moonshot pricing · All comparisons