Gemma 4 31B vs Kimi K2.6
At a Glance
| Compare |  Gemma 4 31BGoogle DeepMind |  Kimi K2.6Moonshot AI |
|---|---|---|
| Intelligence, Cost, and Efficiency | ||
| IntelligenceHigher is better · MM Intelligence v2.5 | UnrankedNot in the 46-model eligible cohort | #32 of 4650.5 score · 2/3 sources · provisional · missing ARC-AGI-2 · full-core range 33.6–67.0 |
| CostLower is better · Published-token output estimate | UnrankedNot in the 44-model eligible cohort | #21 of 44$0.095 per LiveBench case |
| EfficiencyHigher is better · MM Efficiency v1.5 | UnrankedNot in the 38-model eligible cohort | #24 of 3850.1 score · 2/3 sources · provisional · missing ARC-AGI-2 · full-core range 41.7–58.4 |
| Pricing and Limits | ||
| Input priceFrom · USD / 1M tokens | $0.090Openrouter ↗ · Sep 22, 2026 | $0.75Deepinfra ↗ · Sep 22, 2026 |
| Output priceFrom · USD / 1M tokens | $0.34Openrouter ↗ · Sep 22, 2026 | $3.50Deepinfra ↗ · Sep 22, 2026 |
| Context windowMaximum documented tokens | 262K | 262K |
| Model facts checked | Sep 3, 2026View model evidence → | Aug 28, 2026View model evidence → |
Token prices are the lowest available sourced USD rates; input and output may use different providers. Cost ranking estimates output spend on LiveBench, not a full request bill. Ranking methodology →
Available Benchmarks
| Benchmark | Gemma 4 31B | Kimi-K2.6 |
|---|---|---|
| LMArena Document Arenadocument-2026-09-13-d25aabda0010 · arena_rating · leader | 1,424.5198% of row best · rating · gemma-4-31b; 95% CI [1416.71159012, 1432.31286462]; votes 12326; rank 35 | 1,450.74100% of row best · rating · kimi-k2.6; 95% CI [1443.10467262, 1458.37474735]; votes 11291; rank 27 |
| LMArena Text Arenatext-2026-09-13-d25aabda0010 · arena_rating · leader | 1,441.6899% of row best · rating · gemma-4-31b; 95% CI [1434.13929794, 1449.22204224]; votes 5894; rank 64 | 1,454.92100% of row best · rating · kimi-k2.6; 95% CI [1450.40277919, 1459.43693683]; votes 37502; rank 38 |
| LMArena Vision Arenavision-2026-09-13-d25aabda0010 · arena_rating · statistical tie | 1,275.98100% of row best · rating · gemma-4-31b; 95% CI [1269.84273447, 1282.11308989]; votes 35713; rank 33 | 1,280.40100% of row best · rating · kimi-k2.6; 95% CI [1273.38025179, 1287.42397051]; votes 15347; rank 28 |
| Overall ResultCounted from the protocol-matched rows above · 1 tie | 0 benchmark winsNo overall winner | 2 benchmark winsNo overall winner |
Third-party benchmark Only like-for-like primary-publisher results are shown; raw scores, relative scores, configuration, and token spend remain visible.
Side-by-Side Facts
| Field | Gemma 4 31B | Kimi-K2.6 |
|---|---|---|
| Developer | Google DeepMind | Moonshot AI |
| Family | Gemma 4 | Kimi K2 6 |
| Model | Gemma 4 31B | Kimi-K2.6 |
| Version | Gemma 4 31B | Kimi-K2.6 |
| Lifecycle | active | active |
| Released | 2026-03-11 | 2026-04-20 |
| Knowledge cutoff | Unknown | Unknown |
| Input modalities | Text, Image | Text, Image |
| Output modalities | Text | Text |
| Context window | 262K | 262K |
| Total parameters | 31B | 1T |
| Active parameters | Unknown | 32B |
| License | apache-2.0 | other |
| Open weights | Yes | Yes |
| API available | Yes | Yes |
| Self-hostable | Yes | Yes |
| Provider access | Cerebras (Standard), Deepinfra (Standard), Google Gemini (Standard), Openrouter (Standard), Together Ai (Standard) | Deepinfra (Standard), Fireworks Ai (Standard), Hugging Face (Standard), Openrouter (Standard), Together Ai (Standard) |
| Capabilities | chat, generation, reasoning, structured_outputs, tools | chat, generation, reasoning, tools |
Gemma 4 31B Capabilities
Kimi K2.6 Capabilities
Related Comparisons
All image comparisons →| A | Pair | B | Context |
|---|---|---|---|
Gemma 4 31BGoogle DeepMind | vs |  GPT-6 AstraOpenAI | cross-developer peersimage, text |
Claude Fable 5.1Anthropic | vs | Gemma 4 31BGoogle DeepMind | cross-developer peersimage, text |
Gemini 3.1 ProGoogle DeepMind | vs | Gemma 4 31BGoogle DeepMind | developer peerstext |
 DeepSeek-V4.1-FlashDeepSeek | vs | Gemma 4 31BGoogle DeepMind | cross-developer peersimage, text |
Gemma 4 31BGoogle DeepMind | vs | Grok 4.6xAI | cross-developer peersimage, text |
Gemma 4 31BGoogle DeepMind | vs |  Qwen3.8-MaxQwen | cross-developer peersimage, text |
Gemma 4 31BGoogle DeepMind | vs | Kimi-K3Moonshot AI | cross-developer peersimage, text |
 MiniMax-M3MiniMax | vs | Gemma 4 31BGoogle DeepMind | cross-developer peersimage, text |
Gemma 4 31BGoogle DeepMind | vs |  GLM-5.3Z.ai | cross-developer peerstext |
Gemma 4 31BGoogle DeepMind | vs |  Hy4 previewTencent | cross-developer peerstext |
Seed 2.1 ProByteDance Seed | vs | Gemma 4 31BGoogle DeepMind | cross-developer peersimage, text |
Gemma 4 31BGoogle DeepMind | vs |  Mistral Large 3Mistral AI | cross-developer peersimage, text |
Primary Evidence
Sources and Freshness
Questions
Gemma 4 31B vs Kimi K2.6 FAQs
Is Gemma 4 31B or Kimi K2.6 better for coding?+
This comparison does not currently contain a protocol-matched coding benchmark for both Gemma 4 31B and Kimi K2.6, so Model Markets cannot name a coding leader from pricing, context size, or capability labels alone.
Which is cheaper, Gemma 4 31B or Kimi K2.6?+
Gemma 4 31B is $0.090 and Kimi K2.6 is $0.75 per million tokens, so Gemma 4 31B is cheaper on this metric. Gemma 4 31B is $0.34 and Kimi K2.6 is $3.50 per million tokens, so Gemma 4 31B is cheaper on this metric.
Which has a larger context window, Gemma 4 31B or Kimi K2.6?+
Neither model has a larger sourced context window in this comparison. Gemma 4 31B is 262K and Kimi K2.6 is 262K.
Which performs better in benchmarks, Gemma 4 31B or Kimi K2.6?+
There is no overall benchmark winner: An overall winner requires at least two decisive benchmarks from at least two original publishers.
Can Gemma 4 31B or Kimi K2.6 be self-hosted?+
Both models have the same recorded self-hosting status: supported. Gemma 4 31B is open weight; Kimi K2.6 is open weight.
Can Gemma 4 31B and Kimi K2.6 understand images?+
Gemma 4 31B is documented with image input; Kimi K2.6 is documented with image input. This reflects supported input modalities, not vision quality.
Which can generate longer answers, Gemma 4 31B or Kimi K2.6?+
Neither has a larger sourced maximum output. Gemma 4 31B is — and Kimi K2.6 is —.
Do Gemma 4 31B and Kimi K2.6 support reasoning and tool use?+
Gemma 4 31B: reasoning, tool calling, and image input. Kimi K2.6: reasoning, tool calling, and image input. Feature support does not establish relative quality.
Which is available from more inference providers, Gemma 4 31B or Kimi K2.6?+
Gemma 4 31B has 5 sourced provider routes; Kimi K2.6 has 5, a tie.
Which offers better value, Gemma 4 31B or Kimi K2.6?+
There is no universal value winner. Compare the input and output prices above with the matched benchmark result for your workload: cheaper tokens can be offset by different quality, token usage, latency, or provider availability.