llmrun
Knowledge

LiveBench 2026 Research Leaderboard

LiveBench 2026 Research measures finding and combining information to answer research questions. It belongs to the 2026 LiveBench release (first snapshot September 2, 2026), a redesigned LiveBench in which each of seven categories is scored by a single task and questions are refreshed to limit training-set contamination. It is a different test from the original LiveBench boards, so the scores are not comparable.

Source: livebench15 open models ranked+33 proprietaryData through Sep 2026

Open modelsAll models

All models ranked on LiveBench 2026 Research

Proprietary / closed models are shown dimmed — you can't run them locally, but they show where the open field stands.

#ModelScoreVRAM @ Q4
1Claude Opus 5 · proprietary
88.9
API only
2Muse Wonder 1.0 · proprietary
88.0
API only
3Claude Fable 5 · proprietary
86.9
API only
4Grok 4.7 · proprietary
86.4
API only
5Grok 4.5 · proprietary
85.9
API only
6DeepSeek V4.1 Flash · 763.2B
85.3
~458 GB
7GPT 6 Astra · proprietary
85.2
API only
8Claude Opus 5.5 · proprietary
85.0
API only
9Grok 4.6 · proprietary
84.5
API only
10Claude Sonnet 5.5 · proprietary
83.6
API only
11GPT 6.1 Sol · proprietary
83.5
API only
12Gemini 4 Argon · proprietary
83.1
API only
13Claude Sonnet 5 · proprietary
82.8
API only
14Claude Fable 5.1 · proprietary
82.0
API only
15Claude Opus 4.8 · proprietary
81.8
API only
16Muse Spark Open 816B A42B · 822.0B
80.3
~543 GB
17Gemini 3.8 Flash · proprietary
79.1
API only
18Kimi K3 · 2779.9B
79.0
~1674 GB
19MiMo V2.6 Pro RL · 1020B
76.7
~613 GB
20Gemini 3.7 Flash · proprietary
76.5
API only
21Qwen3.8 Flash · proprietary
74.6
API only
22Smaug Flash · 304.2B
74.1
~609 GB
23Claude Haiku 5.5 · proprietary
73.7
API only
24GLM 5.3 · 753.3B
73.2
~456 GB
25Beam 1.0 · proprietary
72.8
API only
26Step 5 · proprietary
72.8
API only
27NVIDIA Nemotron 3.5 Super VL 120B A12B BF16 · 124.5B
70.9
~82 GB
28GLM 5.3 Flash · 321.3B
70.1
~195 GB
29Qwen3.8 Max · proprietary
69.9
API only
30GPT 6 Sol · proprietary
69.3
API only
31DeepSeek V4 Flash Vision Exp · 304.6B
69.2
~183 GB
32Muse Spark 1.3 · proprietary
68.6
API only
33MiniMax M3 · 427.0B
68.0
~257 GB
34Muse Spark 1.2 · proprietary
67.7
API only
35Gemini 3.1 Pro · proprietary
65.7
API only
36Claude Sonnet 4.6 · proprietary
63.6
API only
37GPT 5.5 · proprietary
61.6
API only
38GPT 5.6 Terra · proprietary
60.4
API only
39Inkling · 952.4B
59.9
~572 GB
40Qwen3.8 27B · 27.8B
59.4
~17 GB
41GPT 5.6 Sol · proprietary
54.5
API only
42GPT 5.6 Luna · proprietary
51.0
API only
43Gemma 4 31B IT · 31.3B
44.4
~20 GB
44GPT 6 Luna · proprietary
43.5
API only
45NVIDIA Nemotron 3 Ultra 550B A55B BF16 · 560.5B
43.2
~370 GB
46Claude Haiku 4.5 · proprietary
42.2
API only
47NVIDIA Nemotron 3.5 Lightning 30B A3B BF16 · 31.6B
40.0
~19 GB
48Mistral Large 4 · proprietary
28.1
API only

Score vs model size

Which models give the most quality for their size — the ones worth running locally.

100B1Tmodel size (log scale) →85.340.0Muse Spark Open 816B A42B · 822B · 80.3Kimi K3 · 2.8T · 79.0MiMo V2.6 Pro RL · 1T · 76.7GLM 5.3 · 753B · 73.2GLM 5.3 Flash · 321B · 70.1DeepSeek V4 Flash Vision Exp · 305B · 69.2MiniMax M3 · 427B · 68.0Inkling · 952B · 59.9Gemma 4 31B IT · 31B · 44.4NVIDIA Nemotron 3 Ultra 550B A55B BF16 · 561B · 43.2NVIDIA Nemotron 3.5 Lightning 30B A3B BF16 · 32B · 40.0Qwen3.8 27B · 28B · 59.4Qwen3.8 27BNVIDIA Nemotron 3.5 Super VL 120B A12B BF16 · 124B · 70.9NVIDIA Nemotron 3.5 S…Smaug Flash · 304B · 74.1Smaug FlashDeepSeek V4.1 Flash · 763B · 85.3DeepSeek V4.1 Flash
Each dot is a model. Up = higher score, left = smaller (easier to run locally). The dashed line marks the efficiency frontier — the best score you can get at each size or smaller.
  • Qwen3.8 27B, 28B, score 59.4 — on the efficiency frontier (best score at its size or smaller).
  • NVIDIA Nemotron 3.5 Super VL 120B A12B BF16, 124B, score 70.9 — on the efficiency frontier (best score at its size or smaller).
  • Smaug Flash, 304B, score 74.1 — on the efficiency frontier (best score at its size or smaller).
  • DeepSeek V4.1 Flash, 763B, score 85.3 — on the efficiency frontier (best score at its size or smaller).

LiveBench 2026 Research: frequently asked questions

What is the best open LLM on LiveBench 2026 Research?
DeepSeek V4.1 Flash is the top open model on LiveBench 2026 Research, scoring 85.3. Among all models tested — including proprietary ones — it ranks #6. The top model overall is Claude Opus 5 (Anthropic) at 88.9.
What's the best LiveBench 2026 Research model you can run on a 24 GB GPU?
Qwen3.8 27B is the highest-scoring open model that fits in 24 GB at 4-bit quantization (~17 GB), scoring 59.4 on LiveBench 2026 Research.
Can open models match proprietary models on LiveBench 2026 Research?
Not quite on LiveBench 2026 Research: the strongest proprietary model (Claude Opus 5) scores 88.9, ahead of the best open model (DeepSeek V4.1 Flash) at 85.3 — but you can run the open one yourself.

Scores aggregated from livebench. llmrun does not run this benchmark — see the source for methodology, or the about benchmarks for what it measures.