Title card for the Hy4 flash versus DeepSeek V4.1 Flash comparison, showing Hy4 flash at $0.070 input and $0.281 output per million tokens against DeepSeek V4.1 Flash at $0.15 and $0.60, with a footer reading both figures are vendor list prices and the DeepSeek side is its off-peak card.
Guides & Insights

Hy4 flash vs DeepSeek V4.1 Flash: Tencent Picked This Fight Itself

Author

jinhao.song

Date Published

Latest models · 20View all models →
Benchmarks: Artificial Analysis · updated daily
Back to all posts

Tencent chose the opponent. The research post for Hy4 flash, published on 16 November 2026, prices its own per-task cost against exactly one rival model — DeepSeek V4.1 Flash, in its 1017 snapshot, at the maximum effort tier and at DeepSeek's busy-hour rates — and reports one-eighth. That is an unusually specific comparison for a launch post, and it tells you where Tencent thinks the fight is: not against Alibaba's flagship, not against its own Tencent HY4, but against the cheap open-weight model that coding agents have been running on since September. Three weeks of published numbers sit on DeepSeek's side of that comparison. One chart sits on Tencent's.

So the useful question is not which model is better in the abstract. It is whether a 2.1-fold input-price advantage survives contact with the things that differ between the two, and whether the one-eighth per-task figure means what a reader will assume it means. Both are answerable from published material, and the answers point in opposite directions.

What each company actually shipped

Start with the artefacts, because they are not equivalent and the difference explains most of the disagreement downstream.

DeepSeek V4.1 Flash went up on 10 September 2026 under the MIT licence: 48 safetensors shards totalling roughly 510 gigabytes, a technical report titled "Pushing the Limits of KV Cache Compression", a model card with deployment tags, an fp8 variant, and an API with published off-peak and peak cards. A refreshed snapshot, 1017, replaced it on 17 October and now serves under the same API name. The repository carries 1.69 million downloads and 4,978 likes. Artificial Analysis scores it at 32.6 on the Intelligence Index at maximum effort; Epoch AI's Capabilities Index puts the 1017 checkpoint at 159, 19th of the 253 models it tracks.

Hy4 flash went up on 16 November 2026 as a research post with no model card attached: 224 billion parameters total, 64 billion of them Engram parameters, 12 billion active per token, native multimodality, a 1,048,576-token context, and a plan to open-source later. No weights, no licence, no technical report, no independent score, and benchmark results delivered as PNG charts. Artificial Analysis carries Tencent HY4 at 34.6 as of 5 November and has no Hy4 flash entry at all.

Put those side by side and the matchup has an odd shape. The newer model is the one you cannot inspect, and the older one is the one you can download, fine-tune and serve on your own hardware. That asymmetry runs through every section below.

The price card, honestly compared

Both vendors publish list rates, and both publish more than one of them, which is where most comparisons go wrong.

• Input, per million tokens — Hy4 flash $0.070 flat; DeepSeek V4.1 Flash $0.15 off-peak, $0.30 during peak hours.

• Output, per million tokens — Hy4 flash $0.281; DeepSeek $0.60 off-peak, $1.20 peak.

• Cache hit, per million tokens — Hy4 flash $0.003; DeepSeek $0.003 off-peak, $0.006 peak.

• Peak definition — DeepSeek's peak windows are 01:00–04:00 and 06:00–10:00 UTC, Monday to Friday, excluding Chinese public holidays. Everything else is off-peak, which is to say that for a US or European team the cheap card is the one in force for most of its working day.

• Context — Hy4 flash 1,048,576 tokens; DeepSeek V4.1 Flash 1,000,000 tokens with a 384,000-token output ceiling.

• Mainland card — Hy4 flash ¥0.5 / ¥2 / ¥0.02 per million tokens, which is where the dollar figure comes from rather than from a separately published international rate.

Read at the same hour, the gap is 2.1 times on input and 2.1 times on output: $0.070 against $0.15, $0.281 against $0.60. During DeepSeek's peak windows it widens to 4.3 times. On cached input it is a wash off-peak, because both land at $0.003 — and that matters more than the headline, since a coding agent that re-sends a system prompt, a tool schema and a file tree on every step is paying the cache rate most of the time. On the workload shape this matchup is actually about, the two models are priced identically.

That last line is the one that defuses most of the argument. Tencent's one-eighth is not a cache-heavy comparison; it is a per-task cost on WorkIndex, which folds in output length, retry rate and step count alongside the rate card. Two of those three inputs are things Tencent controls through its own post-training, and it reports the control as a single percentage: output length down 50%.

Generated comparison card listing three rate rows for Hy4 flash and DeepSeek V4.1 Flash: input at $0.070 against $0.15 off-peak and $0.30 peak, output at $0.281 against $0.60 and $1.20, and cache hit at $0.003 against $0.003 and $0.006, with a footer noting the DeepSeek card doubles during its weekday peak windows.

Why 12 billion active parameters is the whole architecture argument

DeepSeek V4.1 Flash is a 552-billion-parameter backbone with 16 billion active per token, built on what DeepSeek calls a Causal Encoder-Decoder design: 40 layers split evenly into a 20-layer causal encoder and a 20-layer decoder, with the technical report's emphasis on KV cache compression rather than on perception. Hy4 flash activates 12 billion per token against a 224-billion total, which is three-quarters of DeepSeek's active count and 40% of its total size.

Tencent's account of where the difference comes from is specific about mechanism and silent about structure. Hybrid attention, LatentMoE and the Engram parameters produce a KV cache one twenty-fourth the size of Hy3's and one sixth of Tencent HY4's. Self-distillation from Tencent HY4, then reinforcement learning, cut end-to-end output length by half. A megakernel with four-bit weights and a DFly decoding path keeps decode above 150 tokens per second. What is not published is anything you could falsify from a checkpoint: layer count, expert count, routing width, vision encoder size, and whether multimodality is native or attached.

The Engram parameters deserve a sentence of their own, because 64 billion of them inside a 224-billion-parameter model is not a normal arrangement. Engram, in Tencent's usage, is a memory-style component holding learned associations rather than a routed expert that runs in every forward pass. The practical consequence is that 29% of the model's total size does not get charged against per-token compute, which is a legitimate way to buy capability per parameter and an illegitimate way to compare parameter counts. A 224B model with 64B of Engram is not smaller than a 552B model in the way the two numbers suggest; it is differently shaped.

DeepSeek's design is checkable and Tencent's is not, and that is not a small difference in a head-to-head. Anyone can download 510 gigabytes of weights and count the layers. Nobody can download Hy4 flash and do anything at all.

The one-eighth, traced

Tencent's per-task comparison is worth following closely because it is the only head-to-head number either company has published for this pair, and it is more careful than launch-post arithmetic usually is.

On WorkIndex — Tencent's internal eight-suite set covering slide creation, slide revision, human-machine co-writing, Markdown co-writing, document editing, table repair, audio and video processing, and visual understanding — Tencent reports Hy4 flash at one-thirteenth the per-task cost of Hy4 preview, one-twentieth of Tencent HY4, and one-eighth of DeepSeek-V4.1-Flash-1017 at maximum effort tier and busy-hour price. Every qualifier in that sentence is Tencent's, and the last two are the ones to hold onto: maximum effort means DeepSeek's thinking budget was set high, and busy-hour means the $0.30/$1.20 card rather than the $0.15/$0.60 one.

Take the qualifiers away and the arithmetic changes a lot. At off-peak rates with the same output lengths, the per-token gap is 2.1 times, not 8. If Hy4 flash also emits half the output tokens, per-task cost roughly halves again on top of the price difference, which puts the two models somewhere between 4 and 5 times apart on a task Tencent designed — and DeepSeek's own published behaviour, controlled thinking effort and a 384,000-token output ceiling, suggests its maximum tier is doing considerably more work per task than a model tuned to answer in half the tokens.

The production numbers Tencent volunteered are the more interesting half of its release, because they come with counters attached rather than a single score. In WeChat Pay's agent flows, against DeepSeek-V4.1-Flash-1017, Tencent reports parity or better across four process stages with about 6% fewer execution steps and 29% shorter single outputs. In WeChat Pay's static scanning, 37% less time to complete a scan and 46% fewer tokens at equal detection quality. In Honor of Kings root-cause analysis, 22% to 39% fewer requests and 40.5% to 61.2% fewer tokens than DeepSeek and GLM-5.3-Flash — while making 51.2% and 7.7% more tool calls than each. In Marvis, task satisfaction up 6.2% on app installation, 4.4% on system setup and diagnostics, 6.9% on document processing and research reports.

Two things are visible in that list. The token and step reductions are large and consistent, which is what a 12B-active model trained to shorten its answers should produce. And the tool-call increase is the honest footnote: Hy4 flash gets there with more, shorter round trips. On a per-request billing model that is a wash or a win; on a per-step wall-clock budget it is a different trade, and Tencent's own framing — that it is 88% faster in WeCom — is a latency claim rather than a work claim.

Evidence asymmetry, which is the actual decision

Here is what the two models can tell you about themselves.

DeepSeek V4.1 Flash has a public harness score, a technical report you can read, weights you can run, and three months of the inference stack meeting the model: quantizations, serving recipes, and a documented set of quirks that other people have already hit. Its benchmark position is not spectacular — 90.6 on Terminal-Bench 2.1, 74.2 on DeepSWE v1.1, 65.4 on NL2Repo-Bench at release, and a 32.6 Intelligence Index — but it is a position someone else measured.

Hy4 flash has a vendor chart, a vendor blind test of 85 employees, a vendor internal benchmark, and a sentence about future open weights. The claims are plausible and internally consistent: a KV cache a sixth the size of the flagship's, output length halved, decode above 150 tokens per second. None of them has been reproduced, and the one that would be easiest to check — whether 12 billion active parameters really does office work as well as DeepSeek's 16 billion — is exactly the one nobody outside Tencent can run, because there are no weights.

That asymmetry does not make Hy4 flash worse. It makes it unproven, and the difference between worse and unproven is where evaluation budget should go. A team that needs a cheap multimodal model it can fine-tune, self-host or audit should stop reading here: DeepSeek V4.1 Flash is the answer, and the 2.1-fold price difference is the cost of that option. A team whose workload is documents, slides, extraction and short agent loops, and which pays per token rather than per task, has a real reason to look at the Tencent card — a reason strong enough that the missing evidence is worth buying around rather than waiting for.

Running the comparison on one key

Neither Hy4 flash nor Tencent's own roadmap makes this easy: as of 16 November, Hy4 flash is served through Tencent Cloud's TokenHub and Tencent's consumer and enterprise products, with third-party platforms announced but undated, and DeepSeek V4.1 Flash is on DeepSeek's own API and on open weights.

The half of this matchup you can run today is the half that sets the bar. deepseek/deepseek-v4.1-flash is on OrcaRouter at $0.15 in and $0.60 out per million tokens with cache reads at $0.003 — DeepSeek's own off-peak list price with nothing added — over a 1,048,576-token context, and the 1017 snapshot sits at the same rate under its own identifier so you can pin the checkpoint Tencent benchmarked against rather than the alias. The Tencent route on the same key is tencent/hy4-preview at $0.834 and $2.501, which is Hy4 flash's teacher; running your own task set on preview and on DeepSeek gives you the teacher-to-rival ratio now, and the flash tier can be slotted into the same harness when it lands.

Screenshot of the OrcaRouter model page for deepseek/deepseek-v4.1-flash in English, captured 16 November 2026, showing the model name, the route identifier, a 1,048,576-token context and the $0.15 input / $0.60 output per million token pricing panel.

That is the pattern worth copying from Tencent's own release. Its argument is a per-task cost on eight task types it chose. Yours should be a per-task cost on the tasks you actually pay for, measured on both models, with tokens, steps and wall-clock recorded — and the three of those together are what will tell you whether a 2.1-fold rate advantage turns into anything at the invoice.

Automatic failover is the unglamorous part of that plan. A model in its first week, served from a promotion with no stated end date, is exactly the model you want a fallback chain behind rather than a migration onto; the same key already reaches DeepSeek, so a rule that retries the incumbent before the response starts costs a config line and removes the launch-week risk from the experiment.

Screenshot of the OrcaRouter model page for tencent/hy4-preview in English, captured 16 November 2026, showing the route identifier, a 64K maximum output, $0.834 input and $2.501 output per million tokens, a $0.042 cache read rate and a measured 39.0 tokens per second.

What would settle it

Three things, in order of how much they would change the answer.

An independent score for Hy4 flash. Artificial Analysis indexed Tencent HY4 eight days after it shipped; a flash entry would at least put both models on one harness, and the 32.6-to-34.6 gap between DeepSeek's flash and Tencent's flagship is the band this matchup is arguing about. Nobody should be surprised if a distilled model lands below its teacher, and the size of that drop is the single most useful unknown in this comparison.

Weights, or a date for them. Tencent's own August preview shipped Apache 2.0 with vLLM and SGLang recipes on day one, so the company knows how to do this. If Hy4 flash follows, its efficiency claims become checkable and the 510-gigabyte download stops being the only option in the room.

A neutral per-task benchmark. Tencent's one-eighth is a good-faith number with two qualifiers attached and a harness nobody else can run. Reproduce it on public agent tasks and the argument is over in an afternoon, one way or the other.

Until then the honest summary is narrower than either vendor's: DeepSeek V4.1 Flash is the cheap multimodal model with an audit trail, Hy4 flash is the cheaper one without, and the price difference is smaller on cache-heavy work than on a rate card. That is not a verdict. It is an accurate description of what 16 November actually changed, which was a rate card and a plan — and a fight Tencent chose on a scoreboard it drew itself.

Compared in this article4

Detected from this article · Benchmarks: Artificial Analysis · updated daily