Sarvam 30B
- Input
- $0.026/M tok
- Output
- $0.11/M tok
- Blended
- $0.047/M tok · 3:1
- Speed
- ~110tok/sest.
- Context
- 66Ktokens
Long-tail import from Artificial Analysis (2026-09-24); AA variant: Sarvam 30B (high). Price = AA median host price.
Rankings
Self-hostingEstimated
How we estimate →What it would cost to run these open weights yourself on rented GPUs, compared with the API price above.
- Hardware
- 1× H200 141GB
- FP8 weights
- Throughput
- ~9,509 tok/s
- many requests batched
- Price
- $0.040–0.33 /M tok
- busy → light use
- On your own machine
- RTX 5090 32GB (int4, 300+ tok/s single-stream)
- single consumer GPU or Mac
Assumptions (5)
- FP8 weights (30B params, 2.4B active per token) + 25% KV-cache headroom ≈ 38 GB
- 1× H200 141GB at $3.59–$7.91/GPU-hour on-demand (2026-09-24)
- ~9,509 output tok/s aggregate at batch 256 (bandwidth-bound); MoE compute scales with active params
- Blended 3:1 input:output; prefill ~194,808 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.
Writing & chat
IFBenchVerified
#127 of 13926.5%
- #1MiniMax-M382.9%
- #2Qwen3.8 2.4T A95B82.8%
- #126MiniCPM-V 4.6 1.3B26.7%
- #127Sarvam 30B26.5%
- #128Granite 4.0 H 1B26.2%
artificialanalysis.ai · 2026-09-24
MMMLUEstimated
~last of 5 (estimated)below measured range (<81.3%)
No published score — estimated from related benchmarks
LMArena Text, Non-English (Elo)Estimated
~#48 of 57 (estimated)~1314 Elo
- #1Gemini 4 ArgonFew votes1512
- #2Claude Opus 5.51497
- #47INTELLECT-3Few votes1327
- ~#48Sarvam 30BEstimated1314
- #48Llama 4 Maverick1311
No published score — estimated from related benchmarks
LMArena Text (Elo)Estimated
~#49 of 57 (estimated)~1338 Elo
- #1Gemini 4 Argon1525
- #2Claude Opus 5.51507
- #48Llama 3.1 Nemotron Ultra 253B v1Few votes1348
- ~#49Sarvam 30BEstimated1338
- #49Olmo 3.1 32B Instruct1330
No published score — estimated from related benchmarks
EQ-Bench Creative Writing v3 (Elo)Estimated
~#22 of 26 (estimated)~1415 Elo
- #1GPT-6 Astra2173
- #2Claude Fable 5.12162
- #21Gemini 3.1 Pro1491
- ~#22Sarvam 30BEstimated1415
- #22Mistral Large 31412
No published score — estimated from related benchmarks
Hard reasoning
Humanity's Last ExamVerified
#130 of 1987.5%
- #1Claude Fable 5.274.3%
- #2Claude Opus 5.561.4%
- #129Qwen3 Next 80B A3B Instruct7.6%
- #130Sarvam 30B7.5%
- #131LFM2.5-24B-A2B7.4%
artificialanalysis.ai · 2026-09-24
AIME (latest)Estimated
~#23 of 34 (estimated)~53.0%
- #1GLM-5.3 NVFP499.5%
- #2Beam 1.0 501B A23B97.7%
- #22Phi-4-reasoning-plus78.0%
- ~#23Sarvam 30BEstimated53.0%
- #23Mistral Large 338.0%
No published score — estimated from related benchmarks
AA Intelligence IndexEstimated
~#90 of 103 (estimated)~9.2
- #1Gemini 4 Carbon59.4
- #2Claude Opus 5.557.6
- #89Qwen3 Coder Next9.2
- ~#90Sarvam 30BEstimated9.2
- #90Granite 4.2 3B9.1
No published score — estimated from related benchmarks
Agents
τ²-bench (Telecom)Verified
#58 of 11934.5%
- #1Step 3.7 Flash98.5%
- #2DeepSeek-V4-Pro-0813-nvfp4-DSpark98.3%
- #57North Mini Code37.4%
- #58Sarvam 30B34.5%
- #59Gemma 4 12B31.9%
artificialanalysis.ai · 2026-09-24
OSWorld-VerifiedEstimated
~#12 of 15 (estimated)~53.6%
- #1Kimi K384.8%
- #2Qwen3.8 27B84.3%
- #11Muse Glimmer65.9%
- ~#12Sarvam 30BEstimated53.6%
- #12Nemotron 3.5 Super VL NVFP452.3%
No published score — estimated from related benchmarks
OSWorld 2.0Estimated
~#3 of 5 (estimated)~53.8%
No published score — estimated from related benchmarks
BrowseCompEstimated
~#21 of 22 (estimated)~42.2%
- #1Atria Dawn Preview92.5%
- #2GPT-6 Astra91.5%
- #20Nemotron 3 Ultra44.4%
- ~#21Sarvam 30BEstimated42.2%
- #21Mistral Small 421.3%
No published score — estimated from related benchmarks
GDPval-AA (Elo)Estimated
~#54 of 92 (estimated)~777 Elo
- #1Claude Opus 5.51866
- #2Muse Wonder Full1857
- #53Qwen3.5 397B A17B780
- ~#54Sarvam 30BEstimated777
- #54Gemini 3.1 Pro776
No published score — estimated from related benchmarks
AutomationBench-AAEstimated
~#73 of 85 (estimated)~0.7%
- #1Claude Fable 5.276.4%
- #2Muse Wonder Full71.3%
- #72Nemotron 3.5 Lightning0.8%
- ~#73Sarvam 30BEstimated0.7%
- #73Ministral 3 3B0.7%
No published score — estimated from related benchmarks
τ-Bench Banking (AA)Estimated
~#72 of 91 (estimated)~6.8%
- #1Muse Spark 1.350.5%
- #2GLM-5.350.3%
- #70Qwen3.5 9B7.0%
- ~#72Sarvam 30BEstimated6.8%
- #72Ministral 3 14B6.6%
No published score — estimated from related benchmarks