SAVRN
Search Contact SAVRN

SAVRN Model Hub · Comparisons

DeepSeek-V4-Flash-0731 vs DeepSeek-V4-Flash-DSpark

DeepSeek-V4-Flash-0731 has 304.2B parameters and DeepSeek-V4-Flash-DSpark has 165.3B parameters; both are released under MIT License; at 16-bit, DeepSeek-V4-Flash-0731 needs about 730 GB (3x MI325X from $6.00 an hour) and DeepSeek-V4-Flash-DSpark about 396.6 GB (2x MI325X from $4.00 an hour).

Published metadata for 2 models, each read from its own repository.
Field DeepSeek-V4-Flash-0731
deepseek-ai/DeepSeek-V4-Flash-0731
DeepSeek-V4-Flash-DSpark
deepseek-ai/DeepSeek-V4-Flash-DSpark
Publisher DeepSeek DeepSeek
Task Text generation Text generation
Modality Text Text
Parameters, as reported 304.2B parameters 165.3B parameters
Architecture DeepseekV4ForCausalLM DeepseekV4ForCausalLM
Library transformers transformers
Context length 1,048,576 tokens 1,048,576 tokens
Repository size 166.9 GB 166.9 GB
Artifact formats safetensors safetensors
License mit mit
Access Open weights, no gate Open weights, no gate
Memory at 16-bit (weights and margin) 730 GB 396.6 GB
Cheapest GPUs at 16-bit, per hour 3x MI325X, $6.00 2x MI325X, $4.00
Memory at 4-bit (weights and margin) 182.5 GB 99.2 GB
Cheapest GPUs at 4-bit, per hour 1x MI300X, $1.85 1x MI300X, $1.85
Revision viewed 7872f01b1d1f 62af8fffb2f7
Downloads reported by the hub 4.3M 1M
Last observed 2026-09-18 2026-09-18

An evaluation row appears only where at least two of these models report the same benchmark with the same stated configuration, metric, unit and setup. Different evaluators stay named in each cell. Values are shown as reported: no unit conversion, no ranking.

Other Reported Results

These results are listed for each model on its own, because the conditions needed to compare them are not stated or do not match. Two results that leave a condition blank are not assumed to share it.

DeepSeek-V4-Flash-0731

BenchmarkConditionsResultReported byRevisionDate
datacurve/deep-swe Task deep_sweMetric deep_sweComparison conditions not established 54.4 Model Card
Reported by a third party
Evaluated revision not stated 2026-08-03
harborframework/terminal-bench-2.1 Task terminalbench_2_1Metric terminalbench_2_1Setup DeepSeek Harness (minimal mode), max reasoning effort, temperature=1.0, top_p=0.95.Comparison conditions not established 82.7 Model Card
Reported by a third party
Evaluated revision not stated 2026-08-01
hkust-nlp/Toolathlon Task toolathlon_verifiedMetric toolathlon_verifiedComparison conditions not established 70.3 deepseek-ai/DeepSeek-V4-Flash-0731 model card
Reported by a third party
Evaluated revision not stated 2026-08-01

SAVRN's Notes on DeepSeek-V4-Flash-0731

Only 6 of the 256 routed experts fire on any given token, but all 304.2B parameters have to be resident, and that sets the bill for DeepSeek-V4-Flash-0731. At 16-bit the working set is 730 GB: three MI325X cards with 256 GB each at $6.00 an hour. At 8-bit, 365 GB fits on two MI300X at $3.70; at 4-bit, 182.5 GB fits on one MI300X at $1.85. It generates text over a 1,048,576-token context.

SAVRN Index host prices run from $0.06 in and $0.18 out per million tokens at DeepInfra to $0.44 and $1.32 at Novita, more than a seven-fold spread, so price a run yourself. MIT keeps that simple: commercial use, modification and redistribution, provided the copyright and permission notice travels with the files. Decide how much of the million-token window you will run, since the memory figures are quoted on the weights, and read arXiv:2606.19348, the paper behind it.

SAVRN's Notes on DeepSeek-V4-Flash-DSpark

The publisher says it outright: DSpark is not a new model. It is the DeepSeek-V4-Flash checkpoint with a speculative decoding module attached, so you are evaluating a serving path, not fresh weights. Those weights run 165.3 billion parameters across 256 routed experts with 6 active per token. Precision picks the hardware: 16-bit needs 396.6 GB and two MI325X cards at $4.00 per hour, 8-bit needs 198.3 GB on one MI325X at $2.00, and 4-bit needs 99.2 GB on one MI300X at $1.85.

MIT covers it: commercial use, modification and redistribution with the notices kept. Two checks before buying cards. The memory figures are the floor; a request that uses the 1,048,576-token context adds cache on top, sized from the single key/value head at 512 dimensions. And the speculative decoding path is the point of this release, so confirm your serving stack runs the publisher's inference example.

Questions

Which is larger, DeepSeek-V4-Flash-0731 or DeepSeek-V4-Flash-DSpark?

DeepSeek-V4-Flash-0731 (304.2B parameters) is larger than DeepSeek-V4-Flash-DSpark (165.3B parameters), by the parameter counts their publishers report.

Which is cheaper to run, DeepSeek-V4-Flash-0731 or DeepSeek-V4-Flash-DSpark?

At 4-bit, DeepSeek-V4-Flash-0731 fits on 1x MI300X from $1.85 an hour and DeepSeek-V4-Flash-DSpark on 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

Can I use DeepSeek-V4-Flash-0731 commercially?

Yes. DeepSeek-V4-Flash-0731 is released under MIT License. The MIT License is a short permissive license. It permits commercial use, modification and redistribution, provided the copyright notice and permission notice are included.

Can I use DeepSeek-V4-Flash-DSpark commercially?

Yes. DeepSeek-V4-Flash-DSpark is released under MIT License. The MIT License is a short permissive license. It permits commercial use, modification and redistribution, provided the copyright notice and permission notice are included.

Related Comparisons