Fastest LLM (October 2026)
As of 12 Oct 2026, the fastest good LLM on independent benchmarks is Muse Spark 1.2 (xhigh) by Meta (output speed 325 tok/s), ahead of Grok 4.20 Multi-Agent (output speed 315 tok/s) and Gemini 3.7 Flash (medium) (output speed 285 tok/s).
Output speed in tokens per second, among models that stream at least 100 tokens per second and score above 55 on the BenchLeader Index, so the list is fast models worth using rather than the fastest small ones.
28 models qualify · data as of 12 Oct 2026 · rebuilt once a day (the run starts at 22:03 UTC)
Muse Spark 1.2 xhigh#1 here: output speed 325 tok/sBest value
Ling 3.0 Flash VL most index points per dollar among the top 15 (55.7 at $0.115/M)Best open weights
DeepSeek V4.1 Flash maxhighest-ranked model here you can run yourself (#5)| # | Model | Index | Price $/M | Speed tok/s |
|---|---|---|---|---|
| 1 | Muse Spark 1.2xhighMeta | 62.7 | $2.00 | 325 |
| 2 |  Grok 4.20 Multi-AgentSpaceXAI | 58.8 | $1.56 | 315 |
| 3 |  Gemini 3.7 FlashmediumGoogle | 63.3 | $1.50 | 285 |
| 4 | Deepseek v4 Flash VisionmaxDeepSeek | 58.4 | $0.660 | 221 |
| 5 | DeepSeek V4.1 FlashmaxDeepSeek | 61.8 | $0.525 | 217 |
| 6 | Gemini 3 FlashthinkingGoogle | 59.2 | $1.13 | 208 |
| 7 | Gemini 3.5 FlashmediumGoogle | 63.0 | $3.38 | 206 |
| 8 | Muse Spark 1.1Meta | 63.1 | $2.00 | 202 |
| 9 |  Qwen3 7maxAlibaba | 59.9 | $3.75 | 197 |
| 10 | Gemini 3.6 FlashhighGoogle | 60.4 | $1.50 | 197 |
| 11 | Muse Spark 1.3maxMeta | 67.7 | $2.00 | 175 |
| 12 |  Claude Haiku 5.5highAnthropic | 61.3 | $0.200 | 174 |
| 13 | Inkling SmallThinking Machines | 55.1 | $0.525 | 160 |
| 14 | Ling 3.0 Flash VLAnt Group | 55.7 | $0.115 | 145 |
| 15 |  Nemotron 3 Ultra 550B A55BthinkingNVIDIA | 56.6 | $1.07 | 142 |
Full ranked list (28 models) · Best for voice agents, with the weighting stated · Methodology
Fastest LLM: questions
- What is the fastest good LLM in October 2026?
- As of 12 Oct 2026, the fastest good LLM on independent benchmarks is Muse Spark 1.2 (xhigh) by Meta (output speed 325 tok/s), ahead of Grok 4.20 Multi-Agent (output speed 315 tok/s) and Gemini 3.7 Flash (medium) (output speed 285 tok/s). BenchLeader reconciles 210 benchmarks from 42 benchmark publishers and rebuilds the ranking once a day (the run starts at 22:03 UTC).
- Which is the best value?
- Ling 3.0 Flash VL gives the most BenchLeader Index points per dollar among the top 15 here: 55.7 at $0.115 per million tokens blended, against $2.00 for Muse Spark 1.2.
- What is the best open-weights option?
- DeepSeek V4.1 Flash (max) by DeepSeek is the highest-ranked open-weights model on this list, at #5.
- How is this ranked?
- Output speed in tokens per second, among models that stream at least 100 tokens per second and score above 55 on the BenchLeader Index, so the list is fast models worth using rather than the fastest small ones. Every score comes from an independent publisher and links to its source on the model page; self-reported launch numbers are excluded.
Cite as: BenchLeader, “Fastest LLM (October 2026)”, https://www.benchleader.com/best/fastest-llm, data as of 12 Oct 2026.