SAVRN Model Hub · Comparisons
DeepSeek-V4-Flash-0731 vs MiniMax-M2.7
DeepSeek-V4-Flash-0731 has 304.2B parameters and MiniMax-M2.7 has 228.7B parameters; DeepSeek-V4-Flash-0731 is released under MIT License and MiniMax-M2.7 under other; at 16-bit, DeepSeek-V4-Flash-0731 needs about 730 GB (3x MI325X from $6.00 an hour) and MiniMax-M2.7 about 548.9 GB (2x MI355X from $5.18 an hour).
| Field | DeepSeek-V4-Flash-0731 deepseek-ai/DeepSeek-V4-Flash-0731 | MiniMax-M2.7 MiniMaxAI/MiniMax-M2.7 |
|---|---|---|
| Publisher | DeepSeek | MiniMax |
| Task | Text generation | Text generation |
| Modality | Text | Text |
| Parameters, as reported | 304.2B parameters | 228.7B parameters |
| Architecture | DeepseekV4ForCausalLM | MiniMaxM2ForCausalLM |
| Library | transformers | transformers |
| Context length | 1,048,576 tokens | 204,800 tokens |
| Repository size | 166.9 GB | 230.2 GB |
| Artifact formats | safetensors | safetensors |
| License | mit | other |
| Access | Open weights, no gate | Open weights, no gate |
| Memory at 16-bit (weights and margin) | 730 GB | 548.9 GB |
| Cheapest GPUs at 16-bit, per hour | 3x MI325X, $6.00 | 2x MI355X, $5.18 |
| Memory at 4-bit (weights and margin) | 182.5 GB | 137.2 GB |
| Cheapest GPUs at 4-bit, per hour | 1x MI300X, $1.85 | 1x MI300X, $1.85 |
| Revision viewed | 7872f01b1d1f | d494266a4aff |
| Downloads reported by the hub | 4.3M | 1.5M |
| 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
| Benchmark | Conditions | Result | Reported by | Revision | Date |
|---|---|---|---|---|---|
| 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 |
MiniMax-M2.7
| Benchmark | Conditions | Result | Reported by | Revision | Date |
|---|---|---|---|---|---|
| SWE-bench/SWE-bench_Multilingual | Task swe_bench_multilingual_%_resolvedMetric swe_bench_multilingual_%_resolvedComparison conditions not established | 76.5 | Model Card Reported by a third party |
Evaluated revision not stated | 2026-08-10 |
| ScaleAI/SWE-bench_Pro | Task SWE_Bench_ProMetric SWE_Bench_ProComparison conditions not established | 56.2 | Model Card Reported by a third party |
Evaluated revision not stated | 2026-04-12 |
| benchflow/skillsbench | Task skillsbench_v1_1Metric skillsbench_v1_1Setup with-skills; BenchFlow harness; OpenHands agent; 87 tasks x 3 trials; full 261/261 coverageComparison conditions not established | 34.9 | SkillsBench v1.1 official leaderboard Reported by a third party |
Evaluated revision not stated | 2026-06-11 |
| claw-eval/Claw-Eval | Task generalMetric generalSetup Pass³% | N=3 | 161 tasksComparison conditions not established | 49.7 | Claw-Eval Leaderboard Reported by a third party |
Evaluated revision not stated | 2026-04-23 |
| claw-eval/Claw-Eval | Task multi_turnMetric multi_turnSetup Pass³% | N=3 | 38 tasksComparison conditions not established | 44.7 | Claw-Eval Leaderboard Reported by a third party |
Evaluated revision not stated | 2026-04-23 |
| harborframework/terminal-bench-2.0 | Task terminalbench_2Metric terminalbench_2Comparison conditions not established | 57 | Model Card Reported by a third party |
Evaluated revision not stated | 2026-04-12 |
| internlm/WildClawBench | Task avg_costMetric avg_costComparison conditions not established | 7.2 | WildClawBench Reported by a third party |
Evaluated revision not stated | 2026-05-22 |
| internlm/WildClawBench | Task avg_timeMetric avg_timeComparison conditions not established | 551 | WildClawBench Reported by a third party |
Evaluated revision not stated | 2026-05-22 |
| internlm/WildClawBench | Task overallMetric overallComparison conditions not established | 33.8 | WildClawBench Reported by a third party |
Evaluated revision not stated | 2026-05-22 |
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 MiniMax-M2.7
Pricing 228.7 billion parameters across 256 experts starts with precision. At 4-bit the model needs 137.2 GB and fits one MI300X with 192 GB at $1.85 per hour; at 8-bit, 274.4 GB pushes you to one MI355X with 288 GB at $2.59; full 16-bit takes 548.9 GB and two MI355X at $5.18. The publisher positions M2.7 for agent harnesses and tool-driven work, and the 204,800-token context is sized for long tool transcripts in one window.
The license is listed as other with no summary, so read the publisher's terms in full before a commercial deployment; nothing here confirms redistribution or fine-tuning rights. Access is open and the download is 230 GB across 151 safetensors files. Weigh $1.85 per hour against renting: DeepInfra lists $0.25 per million input tokens and $1.00 output, Novita $0.30 and $1.20, and the crossover is set by how many tokens you push through the card.
Questions
Which is larger, DeepSeek-V4-Flash-0731 or MiniMax-M2.7?
DeepSeek-V4-Flash-0731 (304.2B parameters) is larger than MiniMax-M2.7 (228.7B parameters), by the parameter counts their publishers report.
Which is cheaper to run, DeepSeek-V4-Flash-0731 or MiniMax-M2.7?
At 4-bit, DeepSeek-V4-Flash-0731 fits on 1x MI300X from $1.85 an hour and MiniMax-M2.7 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.