Ripple-65B Preview
- Input
- —
- Output
- —
- Speed
- ~76tok/sest.
- Context
- 1Mtokens
Open weights (Apache 2.0), 65B dense Transformer pretrained on 12T tokens; light post-training focused on agentic coding. Card headline: 80.52 on Terminal-Bench 2.1, a version our Terminal-Bench 4.0 does not accept, so it is left out. No API price or measured speed yet.
Rankings
Self-hostingEstimated
How we estimate →What it would cost to run these open weights yourself on rented GPUs. No API sells this model, so this is its price on the chart.
- Hardware
- 1× B200 180GB
- FP8 weights
- Throughput
- ~2,559 tok/s
- many requests batched
- Price
- $0.32–2.84 /M tok
- busy → light use
- On your own machine
- Mac Studio M5 Max 128GB (int4, ~13 tok/s single-stream)
- single consumer GPU or Mac
Assumptions (5)
- FP8 weights (65B params) + 40% KV-cache headroom ≈ 91 GB
- 1× B200 180GB at $5.98–$14.24/GPU-hour on-demand (2026-09-24)
- ~2,559 output tok/s aggregate at batch 73 (bandwidth-bound)
- Blended 3:1 input:output; prefill ~16,356 tok/s
- Low = 75% utilization at the low GPU price; high = 20% at the high price
Benchmark scores
~ italic, dashed = no published score yet, estimated from related benchmarks.
Coding
SWE-bench VerifiedLab-reported
#9 of 2076.5%
- #1Claude Opus 596.0%
- #2Claude Sonnet 585.2%
- #7Mistral Medium 3.577.6%
- #9Ripple-65B Preview76.5%
- #10Muse Glimmer76.0%
huggingface.co · 2026-11-04
SWE-bench Pro (Public)Estimated
~#11 of 24 (estimated)~62.2%
- #1Claude Fable 5.293.0%
- #2Claude Opus 5.589.9%
- #10Claude Sonnet 563.2%
- ~#11Ripple-65B PreviewEstimated62.2%
- #11Nex-N2.5-Pro61.2%
No published score — estimated from related benchmarks
DeepSWE v1.1Lab-reported
#38 of 4544.3%
- #1Gemini 4 Argon78.8%
- #2DeepSeek V4.1 Flash 101777.0%
- #37Qwen3.8 Max (0902)51.0%
- #38Ripple-65B Preview44.3%
- #39Beam 1.0 501B A23B43.3%
huggingface.co · 2026-11-04
Terminal-Bench 4.0Estimated
~#24 of 102 (estimated)~29.0%
- #1Kimi K3.1 Pro67.2%
- #2Claude Sonnet 5.563.6%
- #22GLM-5.3 Flash32.8%
- ~#24Ripple-65B PreviewEstimated29.0%
- #24DeepSeek V4.1 Flash26.8%
No published score — estimated from related benchmarks
LiveCodeBenchEstimated
~#5 of 28 (estimated)~89.9%
- #1Qwen3.8-Omni-Flash92.6%
- #2Qwen3.8 Flash-Next91.9%
- #4Ternary Bonsai 2 27B90.1%
- ~#5Ripple-65B PreviewEstimated89.9%
- #5Nemotron 3 Ultra89.0%
No published score — estimated from related benchmarks
SciCodeEstimated
~#20 of 119 (estimated)~57.0%
- #1Claude Fable 5.267.7%
- #2Claude Opus 5.566.9%
- #19Grok 4.757.4%
- ~#20Ripple-65B PreviewEstimated57.0%
- #20Gemini 3.8 Flash56.6%
No published score — estimated from related benchmarks
AA Coding Agent IndexEstimated
~#17 of 26 (estimated)~45.9
- #1Claude Opus 5.561.7
- #2Claude Opus 559.7
- #16DeepSeek V4.1 Flash 101746.3
- ~#17Ripple-65B PreviewEstimated45.9
- #17Muse Spark 1.345.7
No published score — estimated from related benchmarks
Writing & chat
IFBenchLab-reported
#12 of 13979.2%
- #1MiniMax-M382.9%
- #2Qwen3.8 2.4T A95B82.8%
- #11Qwen3.8 27B79.5%
- #12Ripple-65B Preview79.2%
- #13Nova 2.0 Pro Preview79.0%
huggingface.co · 2026-11-04
LMArena Text, Non-English (Elo)Estimated
~#17 of 57 (estimated)~1458 Elo
- #1Gemini 4 ArgonFew votes1512
- #2Claude Opus 5.51497
- #16DeepSeek V4.1 Flash1459
- ~#17Ripple-65B PreviewEstimated1458
- #17GPT-5.6 Terra1456
No published score — estimated from related benchmarks
LMArena Text (Elo)Estimated
~#17 of 57 (estimated)~1469 Elo
- #1Gemini 4 Argon1525
- #2Claude Opus 5.51507
- #16DeepSeek V4.1 Flash1474
- ~#17Ripple-65B PreviewEstimated1469
- #17GPT-5.6 Terra1466
No published score — estimated from related benchmarks
EQ-Bench Creative Writing v3 (Elo)Estimated
~#9 of 26 (estimated)~1978 Elo
- #1GPT-6 Astra2173
- #2Claude Fable 5.12162
- #8Grok 4.72007
- ~#9Ripple-65B PreviewEstimated1978
- #9Muse Spark 1.31906
No published score — estimated from related benchmarks
Hard reasoning
GPQA DiamondLab-reported
#60 of 19285.2%
- #1GPT-6 Astra96.1%
- #2Gemini 3.8 Flash95.3%
- #58Ling 3.0 Flash85.5%
- #60Ripple-65B Preview85.2%
- #61MiMo-V2.584.9%
huggingface.co · 2026-11-04
Humanity's Last ExamEstimated
~#65 of 199 (estimated)~27.0%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #64MiMo-V2.527.2%
- ~#65Ripple-65B PreviewEstimated27.0%
- #65Solar Mini 425.8%
No published score — estimated from related benchmarks
AIME (latest)Lab-reported
#5 of 3395.0%
- #1GLM-5.3 NVFP499.5%
- #2Beam 1.0 501B A23B97.7%
- #4Ternary Bonsai 2 27B95.8%
- #5Ripple-65B Preview95.0%
- #6Muse Glimmer94.7%
huggingface.co · 2026-11-04
AA Intelligence IndexEstimated
~#41 of 103 (estimated)~28.7
- #1Gemini 4 Carbon59.4
- #2Claude Opus 5.557.6
- #40MiniMax-M329.2
- ~#41Ripple-65B PreviewEstimated28.7
- #41Quasar 438B26.7
No published score — estimated from related benchmarks