Intelligence
37.3
37
General reasoning and benchmark headroom.
LimitedKimi K2.5 (Non-reasoning) is a budget-priced text-first model from Kimi with heavy runtime profile, partial context coverage, and the clearest fit around agent workflows / reasoning.
Benchmark blend
Dev workflow signal
N/A
Budget tier
Kimi K2.5 (Non-reasoning) currently reads as a budget text-first option with partially published context and a heavy runtime profile.
Decision Strip
Core buy-side signals stay in one pass. The rest of the page expands only after intelligence, speed, context, and price are clear.
General reasoning and benchmark headroom.
LimitedTTFT 1.42s
SituationalHow much prompt and task state can stay in view.
Unavailable$3.00 output / 1M
EfficientEditorial Profile
Positioning, tradeoffs, and fit are consolidated into one read instead of repeating the same story across separate cards.
The Kimi K2.5 (Non-reasoning) AI model by Kimi.
Kimi text-first profile
Agent workflows / Reasoning with partially published context and heavy runtime.
Efficient spend profile. More comfortable for sustained prompt volume if the capability fit is right.
The available source data suggests a balanced profile rather than one dominant edge.
Budget-friendly input pricing is a strength, but raw capability may vary by workload.
Latency profile is better for deliberate runs than rapid back-and-forth chat.
Current metadata points to a text-first profile rather than a broad multimodal one.
Context limits are only partially published, so long-session planning needs extra validation.
Focused chat, retrieval-augmented flows, and narrower production tasks.
Benchmarks
Only benchmark categories with actual signal are shown. Secondary values stay as simple definitions instead of nested micro-cards.
Broad reasoning, knowledge depth, and flagship benchmark posture.
Software implementation, debugging quality, and coding benchmark signal.
Long-horizon execution quality and interactive benchmark evidence.
Specs & Pricing
Specs stay neutral, pricing gets emphasis through values rather than extra containers. Raw provider internals remain in metadata at the end.
This model is relatively efficient on price. It is the easier fit when sustained prompt volume matters.
Metadata
Verification details remain available, but the page no longer forces them ahead of the editorial read.