Up to date
deepseek-ai MIT out 45 days ago
763B parameters (6 of 384 experts per token), images and text in, text out, 1M tokens of context. 34 providers serve it from $0.020 per million input tokens.
Verified 2026-10-25 Rank #10 in LLMs · score 129.4 method v1 Sources 5 Downloads 30d 1.6M · #79 of 619 tracked
Changed this week: Cheapest input price rose 148%: $0.020 → $0.050 per 1M (OpenRouter).2026-10-24 · 12 new quantized builds on the Hub (137 in all).2026-10-24 · 1 new provider: Together AI.2026-10-23
No drawable SVG in the reply.
Why did the open-weights model cross the road?
To get to the other side without an API key.
______
/ \
/ \
| |
\ /
\________/
| |
| |
/o\
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~The ringer: the same four playful tests for every model, replies shown as received and checked by code. A joke, not a benchmark; never ranked.3 of 4 pass · parasail-deepseek-v41-flash on parasail · run 2026-10-09 · temperature 0 · cost $0.036 · 2026-10 thing: supermarket · all models →
Picked by rule from 26 tracked pieces: the lab's own post, the most shared, a field report, the latest.
DeepSeek launches V4.1-Flash, the smallest model in its new architecture family with native visual understanding and a 552B-parameter MoE design.
WARP is an embeddable C inference engine that runs the full 2.78-trillion-parameter Kimi K3 on a 64 GB MacBook Pro at about 0.6 tokens per second by streaming selected experts directly from disk.
DeepSeek-V4.1 Flash is a 552B multimodal model that pushes KV cache compression to handle long-horizon agent workflows with contexts up to 1 million tokens.
The 3 picks above, plus these 22. Top is the radar's score: discussion, source authority and the lab's own posts, fading with age.
Live hosted endpoints, cheapest first, USD per million tokens. Ties break on output price, then provider name.
32 providers with a live price, from $0.020 to $0.45 per million input tokens; the median is $0.21. One dot per provider, cheapest in teal.
Hugging Face does not say who downloads a model or with what. These are the proxies: API providers listing it, quantized builds, fine-tunes/adapters/merges, and Spaces that load it.
Widely served (34 API providers); heavily run locally (137 quantized builds).
Top Spaces running it: townbox/deepseek-profile-guide ♥ 6 · bagegegegw/DeepSeek-V4.1-Flash-UNCENSORED-Demo ♥ 4 · ArtShumov/Whyx-PROmpTea ♥ 3 · akhaliq/DeepSeek-V4.1-Flash ♥ 2
Weights only, an estimate: leave room for the KV cache. Method
About 458 GB at INT4, 916 GB at FP8, 1.83 TB at BF16: weights only, so leave room for the KV cache. Every expert must be resident, so total parameters count.
Estimate params × bytes per param × 1.2 · how we estimate
Claim-matched tests, graded by code. Receipts, not a score.
{
"providers": [
{
"provider": "parasail",
"endpoint_model": "parasail-deepseek-v41-flash",
"source_url": "https://api.parasail.io/v1/chat/completions",
"ttft_ms_median": 2250,
"ttft_any_ms_median": 352,
"tokens_per_s_median": 258.9,
"runs": [
{
"ttft_s": 2.397,
"ttft_any_s": 0.352,
"tokens_per_s": 257.3,
"latency_s": 2.73,
"tokens_out": 702,
"reasoning_chars": 2440,
"finish": "stop"
},
{
"ttft_s": 2.25,
"ttft_any_s": 0.353,
"tokens_per_s": 258.9,
"latency_s": 2.73,
"tokens_out": 708,
"reasoning_chars": 2409,
"finish": "stop"
},
{
"ttft_s": 1.901,
"ttft_any_s": 0.346,
"tokens_per_s": 295.3,
"latency_s": 2.35,
"tokens_out": 694,
"reasoning_chars": 2423,
"finish": "stop"
}
]
}
],
"note": "Single client, wall clock; tok/s counts every output token including hidden reasoning."
} {
"calls": [
{
"name": "get_weather",
"args": {
"city": "Lisbon",
"date": "2026-10-14"
}
},
{
"name": "check_calendar",
"args": {
"date": "2026-10-14"
}
},
{
"name": "book_table",
"args": {
"restaurant": "Cervejaria Ramiro",
"date": "2026-10-14",
"time": "19:30",
"party_size": 2
}
}
],
"checks": {
"get_weather": true,
"check_calendar": true,
"book_table": true
},
"final": "Done! Here's the summary:\n\n- **Weather (Lisbon, 2026-10-14):** Clear, 22°C high, only 5% rain probability — no rain, so we proceeded.\n- **Calendar:** Free evening slot 19:30–21:30.\n- **Booking:** Cervejaria Ramiro, table for 2, Wednesday 2026-10-14 at 19:30.\n\n**Confirmation code: TBL-4471**",
"fixture": "receipts.errand.v1"
} {
"fields_matched": 80,
"fields_total": 80,
"cost_per_1000_usd": 0.2766,
"parsed": true,
"per_item": [
{
"street": true,
"city": true,
"postcode": true,
"country": true
},
{
"street": true,
"city": true,
"postcode": true,
"country": true
},
{
"street": true,
"city": true,
"postcode": true,
"country": true
},
{
"street": true,
"city": true,
"postcode": true,
"country": true
},
{
"street": true,
"city": true,
"postcode": true,
"country": true
},
{
"street": true,
"city": true,
"postcode": true,
"country": true
},
{
"street": true,
"city": true,
"postcode": true,
"country": true
},
{
"street": true,
"city": true,
"postcode": true,
"country": true
},
{
"street": true,
"city": true,
"postcode": true,
"country": true
},
{
"street": true,
"city": true,
"postcode": true,
"country": true
},
{
"street": true,
"city": true,
"postcode": true,
"country": true
},
{
"street": true,
"city": true,
"postcode": true,
"country": true
},
{
"street": true,
"city": true,
"postcode": true,
"country": true
},
{
"street": true,
"city": true,
"postcode": true,
"country": true
},
{
"street": true,
"city": true,
"postcode": true,
"country": true
},
{
"street": true,
"city": true,
"postcode": true,
"country": true
},
{
"street": true,
"city": true,
"postcode": true,
"country": true
},
{
"street": true,
"city": true,
"postcode": true,
"country": true
},
{
"street": true,
"city": true,
"postcode": true,
"country": true
},
{
"street": true,
"city": true,
"postcode": true,
"country": true
}
],
"sample": [
{
"street": "1600 Amphitheatre Parkway",
"unit": null,
"city": "Mountain View",
"region": "CA",
"postcode": "94043",
"country": "US"
},
{
"street": "10 Downing Street",
"unit": null,
"city": "London",
"region": null,
"postcode": "SW1A 2AA",
"country": "GB"
},
{
"street": "Rua Augusta 274",
"unit": null,
"city": "Lisboa",
"region": null,
"postcode": "1100-053",
"country": "PT"
}
],
"fixture": "receipts.addresses.v1"
} {
"lengths_k": [
8,
32,
64
],
"depths": [
0.1,
0.5,
0.9
],
"cells": [
[
1,
1,
1
],
[
1,
1,
1
],
[
1,
1,
1
]
],
"detail": [
{
"k": 8,
"depth": 0.1,
"recalled": true,
"reply": "amber-falcon-7731",
"latency_s": 1.54,
"finish": "stop"
},
{
"k": 8,
"depth": 0.5,
"recalled": true,
"reply": "cobalt-marten-7828",
"latency_s": 0.6,
"finish": "stop"
},
{
"k": 8,
"depth": 0.9,
"recalled": true,
"reply": "saffron-newt-7925",
"latency_s": 0.64,
"finish": "stop"
},
{
"k": 32,
"depth": 0.1,
"recalled": true,
"reply": "jade-lynx-8022",
"latency_s": 1.57,
"finish": "stop"
},
{
"k": 32,
"depth": 0.5,
"recalled": true,
"reply": "umber-kestrel-8119",
"latency_s": 1.38,
"finish": "stop"
},
{
"k": 32,
"depth": 0.9,
"recalled": true,
"reply": "coral-heron-8216",
"latency_s": 0.82,
"finish": "stop"
},
{
"k": 64,
"depth": 0.1,
"recalled": true,
"reply": "slate-badger-8313",
"latency_s": 2.53,
"finish": "stop"
},
{
"k": 64,
"depth": 0.5,
"recalled": true,
"reply": "ochre-otter-8410",
"latency_s": 2.27,
"finish": "stop"
},
{
"k": 64,
"depth": 0.9,
"recalled": true,
"reply": "teal-ibis-8507",
"latency_s": 1.47,
"finish": "stop"
}
]
} Weights, every dated article, and repo commits on one rail.
Teal is the weights (2026-09-10), amber triangles are 10 dated articles (each links to it), grey ticks are repo commits.
162 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.
From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.
Everything else, folded: the full tables, every source.
| Benchmark | Value | Setting / notes | Reported by | Date |
|---|---|---|---|---|
| author-reported AGIEval (EM) | 83.4 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported BBEH · Reason through hard problems | 27.2 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported BBH (EM) | 86.1 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported BigCodeBench · Write code with agents | 60.6 pass@1 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported C-Eval · Work across languages | 92.1 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported CVBench (EM) | 77.9 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported Codeforces · Write code with agents | 3471 rating | DeepSeek-V4.1-Flash-Instruct (Max reasoning effort) | model author | 2026-10-07 |
| author-reported DROP (F1) | 87.9 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported DocVQA · Extract text from documents | 95.6 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported GPQA Diamond · Reason through hard problems | 90.9 pass@1 | DeepSeek-V4.1-Flash-Instruct (Max reasoning effort) | model author | 2026-10-07 |
| author-reported GSM8K · Reason through hard problems | 93 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported HLE (Pass@1) | 36.8 (39.1†) pass@1 | DeepSeek-V4.1-Flash-Instruct (Max reasoning effort) | model author | 2026-10-07 |
| author-reported HellaSwag · Chat and follow instructions | 87.2 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported HumanEval · Write code with agents | 79.4 pass@1 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported LongBench v2 · Read long documents | 45.2 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported MATH-500 · Reason through hard problems | 61.1 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported MATH-500 · Reason through hard problems | 65.6 pass@1 | DeepSeek-V4.1-Flash-Instruct (Max reasoning effort) | model author | 2026-10-07 |
| author-reported MGSM · Work across languages | 80.2 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported MMLU-Pro · Reason through hard problems | 74.1 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported MMMU-Pro · Understand images and video | 56.5 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported MultiLoKo (LLM-Judge) | 45.5 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported RefCOCO · Understand images and video | 86 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported SimpleQA · Reason through hard problems | 42.3 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported SuperGPQA · Reason through hard problems | 53.1 | DeepSeek-V4.1-Flash-Base | model author | 2026-10-07 |
| author-reported Terminal-Bench · Write code with agents | 90.6 pass@1 | DeepSeek-V4.1-Flash-Instruct (Max reasoning effort) | model author | 2026-10-07 |
| Benchmark | Value | Setting / notes | Reported by | Date |
|---|---|---|---|---|
| author-reported Terminal-Bench · Write code with agents | 30 pass@1 | DeepSeek-V4.1-Flash-Instruct (Max reasoning effort) | model author | 2026-10-07 |
| author-reported Terminal-Bench · Write code with agents | 31.2 pass@1 | DeepSeek-V4.1-Flash-Instruct (Max reasoning effort) | model author | 2026-10-07 |
Labels are evidence, not verdicts: measured a third-party leaderboard row for this model; author-reported a number the model author published in the card or eval files; structural a deterministic signal from the name, config or chat template, or a decision-model estimate. Each chip links to its source. All use cases.
Listed in LLMs
| Pipeline tag | image-text-to-text |
|---|---|
| Library | transformers |
| Architectures | DeepseekV41ForCausalLM |
| Model type | deepseek_v41 |
| Dtype | bfloat16 |
| Params by dtype | BF16: 1,976,441,856 · F32: 42,307,282 · F8_E4M3: 204,015,223,296 · I8: 557,171,343,360 |
| Size band | XXL |
| MoE | yes |
| Experts | 384 |
| Experts per token | 6 |
| Context length | 1,048,576 (config.json max_position_embeddings) |
| License | MIT (mit) |
| Hub snapshot | 2026-10-25 |
vLLM: supported · DeepseekV41ForCausalLM · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · DeepseekV41ForCausalLM · 773cb938fe9484a8f7e7e46fb1b1fe96412c59ed
vLLM registry (7050ef881ff0835f6890a47db1859a24c65481d6) · SGLang registry (773cb938fe9484a8f7e7e46fb1b1fe96412c59ed) · Matched by architecture class name against each project's registry on main; 'supported' means the architecture is registered, not that this checkpoint was tested.
Weights-only memory for DeepSeek-V4.1-Flash: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 1.83 TB · Cluster (≥640 GB) |
|---|---|
| FP8 (1 byte/param) | 916 GB · Cluster (≥640 GB) |
| INT4 (0.5 bytes/param) | 458 GB · Multi-GPU node (<640 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra. MoE: all experts must be resident, so total parameters count.
quantized: 137finetune: 25
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| antirez/deepseek-v4.1-flash-gguf | quantized | gguf | — | — | mit | antirez | 79 | 2.3M |
| dealignai/DeepSeek-V4.1-Flash-UNCENSORED-FP8 | quantized | -FP8 | 763B | XXL | mit | dealignai | 758 | 88.9k |
| pfeifferj/DeepSeek-V4.1-Flash-GSQ-RCO-GGUF | quantized | GGUF | — | — | mit | pfeifferj | 22 | 76.4k |
| vcruz305/DeepSeek-V4.1-Flash-GGUF | quantized | GGUF | — | — | mit | vcruz305 | 44 | 64.4k |
| kyuz0/deepseek-v4.1-flash-gguf | quantized | gguf | — | — | mit | kyuz0 | 4 | 60.5k |
| Lucebox/DeepSeek-V4.1-Flash-ROCMFP23-GGUF | quantized | GGUF | — | — | mit | Lucebox | 10 | 47.2k |
| nvidia/DeepSeek-V4.1-Flash-NVFP4 | quantized | NVFP4 | 763B | XXL | mit | nvidia | 119 | 37.5k |
| smalinin/DeepSeek-V4.1-Flash-GGUF | quantized | GGUF | — | — | mit | smalinin | 3 | 28.5k |
| apetersson/DeepSeek-V4.1-Flash-MixedQ2-GGUF | quantized | GGUF | — | — | mit | apetersson | 5 | 25.4k |
| Jundot/DeepSeek-V4.1-Flash-oQ4e-mtp | quantized | — | 129B | XXL | mit | Jundot | 8 | 14.3k |
| drowzeys/DeepSeek-V4.1-Flash-Abliterated-Cybersecurity-Unleashed | finetune | — | — | — | mit | drowzeys | 212 | 0 |
| TensorFold/DeepSeek-V4.1-Flash-MLX-2bit-MTP | quantized | MLX | 763B | XXL | mit | TensorFold | 5 | 9k |
| pipenetwork/DeepSeek-V4.1-Flash-MLX-mixed-4_8bit | quantized | MLX | 379B | XXL | mit | pipenetwork | 4 | 7.8k |
| mxxm-t/DeepSeek-V4.1-Flash-GGUF | quantized | GGUF | — | — | mit | mxxm-t | 2 | 7.1k |
| inferencerlabs/DeepSeek-V4.1-Flash-MLX-Q4i | quantized | MLX | — | — | — | inferencerlabs | 4 | 6.6k |
| Mia-AiLab/DeepSeek-V4.1-Flash-EXL3-2.9bpw | quantized | EXL3 | 105B | XXL | mit | Mia-AiLab | 22 | 4.9k |
| GreenBitAI/DeepSeek-V4.1-Flash-4bit-paged | quantized | -4bit | 8.78B | M | mit | GreenBitAI | 1 | 5.9k |
| JigSawPT/DeepSeek-V4.1-Flash-DSpark-GGUF | quantized | GGUF | — | — | mit | JigSawPT | 5 | 5.1k |
| Jundot/DeepSeek-V4.1-Flash-oQ3e-mtp | quantized | — | 112B | XXL | mit | Jundot | 5 | 4.8k |
| rapid-mlx/DeepSeek-V4.1-Flash-REAP-2bit-MLX | quantized | MLX | 681B | XXL | mit | rapid-mlx | 1 | 4.8k |
| dealignai/DeepSeek-V4.1-Flash-UNCENSORED-EXL3-2.9bpw | quantized | EXL3 | 105B | XXL | mit | dealignai | 47 | 2.3k |
| OpensourceWTF/DeepSeek-V4.1-Flash-MTPLX-streaming-q2 | quantized | — | 22.7B | L | mit | OpensourceWTF | 4 | 4.1k |
| pipenetwork/DeepSeek-V4.1-Flash-MLX-mixed-4_8bit-engram6 | quantized | MLX | 428B | XXL | mit | pipenetwork | 2 | 4.1k |
| rapid-mlx/DeepSeek-V4.1-Flash-DSpark-4d2e-MLX | quantized | MLX | 7.43B | M | mit | rapid-mlx | 2 | 3.9k |
| Solstice-AI/DeepSeek-V4.1-Flash-GGUF | quantized | GGUF | — | — | mit | Solstice-AI | 2 | 3.7k |
| nanguoyu/DeepSeek-V4.1-Flash-minirun | quantized | — | — | — | mit | nanguoyu | 0 | 3.7k |
| kernelpool/DeepSeek-V4.1-Flash-MXFP4-GGUF | quantized | MXFP4 | — | — | mit | kernelpool | 0 | 3.3k |
| OpensourceWTF/DeepSeek-V4.1-Flash-streaming-repack-exl3-3.0bpw | quantized | exl3 | 22.7B | L | mit | OpensourceWTF | 5 | 2.9k |
| JigSawPT/DeepSeek-V4.1-Flash-GGUF | quantized | GGUF | — | — | mit | JigSawPT | 8 | 2.7k |
| AMAImedia/DeepSeek-V4.1-Flash-1017-FP8-GGUF | quantized | -FP8 | 763B | XXL | mit | AMAImedia | 1 | 2.8k |
| 0xSojalSec/DeepSeek-V4.1-Flash-MLX-MAC | quantized | MLX | 763B | XXL | mit | 0xSojalSec | 0 | 2.7k |
| pokey2025/DeepSeek-V4.1-Flash-ik-llama-GGUF | quantized | GGUF | — | — | mit | pokey2025 | 1 | 2.5k |
| s-zaizen/DeepSeek-V4.1-Flash-Abliterated | finetune | — | 763B | XXL | mit | s-zaizen | 16 | 1.7k |
| orcarouter/DeepSeek-V4.1-Flash-Uncensored-MLX | quantized | MLX | — | — | mit | orcarouter | 48 | 0 |
| pipenetwork/DeepSeek-V4.1-Flash-REAP25-MLX-mixed-4_8bit | quantized | MLX | 311B | XXL | mit | pipenetwork | 0 | 2.3k |
| pipenetwork/DeepSeek-V4.1-Flash-REAP50-MLX-mixed-4_8bit | quantized | MLX | 243B | XXL | mit | pipenetwork | 0 | 2.3k |
| LibertAIDAI/DeepSeek-V4.1-Flash-NVFP4 | quantized | NVFP4 | 386B | XXL | mit | LibertAIDAI | 13 | 1.4k |
| Lucebox/DeepSeek-V4.1-Flash-DSpark-GGUF | quantized | GGUF | — | — | mit | Lucebox | 1 | 1.7k |
| bot-lab-21/DeepSeek-V4.1-Flash-EXL3-3.5bpw-Pollard | quantized | EXL3 | 332B | XXL | mit | bot-lab-21 | 5 | 1.5k |
| amd/DeepSeek-V4.1-Flash-Quark-MXFP4 | quantized | MXFP4 | 484B | XXL | mit | amd | 1 | 1.6k |
| Date | Commit | SHA |
|---|---|---|
| 2026-10-16 | fix(encoding): preserve tool image order and current-turn thinking | 314113f |
| 2026-10-12 | gitignore | 5254448 |
| 2026-10-12 | Delete files inference/__pycache__/ with huggingface_hub | 12b5c3a |
| 2026-10-12 | update | 4dd75a8 |
| 2026-10-12 | Restore each indexer's K cache on incomplete compression steps (#12) | 8b3cec7 |
| 2026-10-01 | Adding chat template (#68) | 2cba9e4 |
| 2026-09-10 | fix(encoding): preserve tool namespaces in prompts and completions | dba1be0 |
| 2026-09-10 | Update README.md & encoding/README.md | df42c10 |
| 2026-09-10 | Update README.md & encoding/README.md | fb2764a |
| 2026-09-10 | Update inference/README.md | 2bc89ac |
| 2026-09-10 | Update inference/README.md | cae409b |
| 2026-09-10 | Upload DeepSeek_V41_Tech_Report.pdf | 53e70b1 |
| 2026-09-10 | Upload DeepSeek_V41_Tech_Report.pdf | 9574840 |
| 2026-09-10 | Upload DeepSeek_V41_Tech_Report.pdf | 97fcff4 |
| 2026-09-10 | init | 517ef62 |
| 2026-09-10 | Add files using upload-large-folder tool | 678ae2a |
| 2026-09-10 | initial commit | 3fb9e06 |
Hosted endpoints observed for DeepSeek-V4.1-Flash, sorted by input price (USD per million tokens). Each row names its observer and date; prices change daily, so check the source before quoting. Cheapest observed input price: $0.02/M.
| Provider | Model ref | Context | Quant | $/M in | $/M out | tok/s | TTFT | Tools | Status | Observed |
|---|---|---|---|---|---|---|---|---|---|---|
| OpenInference openrouter | deepseek/deepseek-v4.1-flash | 1M | fp4 | $0.02 | $1.00 | — | — | yes | live | 2026-10-24 |
| Sail Research openrouter | deepseek/deepseek-v4.1-flash | 1M | fp4 | $0.05 | $0.15 | — | — | yes | live | 2026-10-24 |
| Baidu openrouter | deepseek/deepseek-v4.1-flash | 1M | fp8 | $0.05 | $0.21 | — | — | yes | live | 2026-10-24 |
| Relace openrouter | deepseek/deepseek-v4.1-flash | 1M | unknown | $0.07 | $0.50 | — | — | yes | live | 2026-10-24 |
| Decart openrouter | deepseek/deepseek-v4.1-flash | 1M | fp4 | $0.09 | $0.18 | — | — | yes | live | 2026-10-24 |
| InferenceNet openrouter | deepseek/deepseek-v4.1-flash | 1016K | fp8 | $0.10 | $0.71 | — | — | yes | live | 2026-10-24 |
| Reka openrouter | deepseek/deepseek-v4.1-flash | 1M | unknown | $0.10 | $0.84 | — | — | yes | live | 2026-10-24 |
| DeepInfra hf, deepinfra, openrouter | deepseek-ai/DeepSeek-V4.1-Flash | 1M | fp8 | $0.11 | $0.34 | 87.3 | 438 ms | yes | status -5 | 2026-10-25 |
| AtlasCloud openrouter | deepseek/deepseek-v4.1-flash | 1M | fp8 | $0.14 | $0.54 | — | — | yes | live | 2026-10-24 |
| StreamLake openrouter | deepseek/deepseek-v4.1-flash | 1000K | fp8 | $0.15 | $0.59 | — | — | yes | live | 2026-10-24 |
| CoreWeave openrouter | deepseek/deepseek-v4.1-flash | 1M | fp8 | $0.15 | $0.65 | — | — | yes | live | 2026-10-24 |
| DigitalOcean openrouter | deepseek/deepseek-v4.1-flash | 1M | unknown | $0.17 | $0.66 | — | — | yes | live | 2026-10-24 |
| Alibaba openrouter | deepseek/deepseek-v4.1-flash | 977K | unknown | $0.17 | $0.66 | — | — | yes | live | 2026-10-24 |
| ModelRun openrouter | deepseek/deepseek-v4.1-flash | 1M | fp4 | $0.17 | $0.67 | — | — | yes | live | 2026-10-24 |
| GMICloud openrouter | deepseek/deepseek-v4.1-flash | 1024K | fp8 | $0.18 | $0.72 | — | — | yes | live | 2026-10-24 |
| Novita hf, openrouter | deepseek/deepseek-v4.1-flash | 1M | fp8 | $0.20 | $0.78 | 79.4 | 1093 ms | yes | live | 2026-10-25 |
| Mancer 2 openrouter | deepseek/deepseek-v4.1-flash | 1M | fp8 | $0.20 | $1.00 | — | — | yes | live | 2026-10-24 |
| Fireworks AI hf, openrouter | accounts/fireworks/models/deepseek-v4p1-flash | 1M | unknown | $0.21 | $0.84 | 95.8 | 589 ms | yes | live | 2026-10-25 |
| Phala openrouter | deepseek/deepseek-v4.1-flash | 1M | unknown | $0.21 | $0.84 | — | — | yes | live | 2026-10-24 |
| Wafer openrouter | deepseek/deepseek-v4.1-flash | 1M | mxfp4 | $0.22 | $0.65 | — | — | yes | live | 2026-10-24 |
| Makora openrouter | deepseek/deepseek-v4.1-flash | 1M | fp8 | $0.26 | $1.17 | — | — | yes | live | 2026-10-24 |
| Crusoe openrouter | deepseek/deepseek-v4.1-flash | 1M | fp8 | $0.29 | $1.20 | — | — | yes | live | 2026-10-24 |
| Together AI hf, openrouter | deepseek-ai/DeepSeek-V4.1-Flash | 1M | unknown | $0.30 | $1.20 | 272.6 | 112 ms | yes | live | 2026-10-25 |
| Baseten hf, openrouter | deepseek-ai/DeepSeek-V4.1-Flash | 1M | unknown | $0.30 | $1.20 | 308.9 | 173 ms | yes | live | 2026-10-25 |
| Parasail parasail, openrouter | parasail-deepseek-v41-flash | 1M | fp8 | $0.30 | $1.20 | — | — | yes | serverless | 2026-10-24 |
| Google openrouter | deepseek/deepseek-v4.1-flash | 1M | unknown | $0.30 | $1.20 | — | — | yes | live | 2026-10-24 |
| Nebius openrouter | deepseek/deepseek-v4.1-flash | 1023K | fp8 | $0.30 | $1.20 | — | — | yes | live | 2026-10-24 |
| PrimeIntellect openrouter | deepseek/deepseek-v4.1-flash | 1M | unknown | $0.30 | $1.20 | — | — | yes | live | 2026-10-24 |
| Cohere openrouter | deepseek/deepseek-v4.1-flash | 1M | unknown | $0.30 | $1.20 | — | — | yes | live | 2026-10-24 |
| SiliconFlow openrouter | deepseek/deepseek-v4.1-flash | 1M | fp8 | $0.30 | $1.20 | — | — | yes | live | 2026-10-24 |
| Modal openrouter | deepseek/deepseek-v4.1-flash | 1M | unknown | $0.30 | $1.20 | — | — | yes | live | 2026-10-24 |
| Venice openrouter | deepseek/deepseek-v4.1-flash | 977K | fp8 | $0.30 | $1.20 | — | — | yes | live | 2026-10-24 |
| Morph openrouter | deepseek/deepseek-v4.1-flash | 1M | fp8 | $0.45 | $0.60 | — | — | yes | live | 2026-10-24 |
| Featherless hf | deepseek-ai/DeepSeek-V4.1-Flash | — | — | — | — | — | — | — | live | 2026-10-25 |
Ranked by source signals (HN, Lobsters, authority domains, official posts) and recency. Summaries are generated from a stored copy of the article text; every sentence cites a span of that text, and nothing is shown that failed the citation check.
Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:31 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:45 UTC · Summary: 2026-10-07 14:59 UTC · Last verified 2026-10-25
Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.