-
Notifications
You must be signed in to change notification settings - Fork 27
fix: Gemma-4 QuantizedKVCache + kv_bits API + Test 9 (mlx-swift-lm b440) #73
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Merged
Merged
Changes from all commits
Commits
Show all changes
4 commits
Select commit
Hold shift + click to select a range
c1b90f1
feat: Gemma-4 QuantizedKVCache fix + Test 9 regression (mlx-swift-lm …
github-actions[bot] f007b3b
docs+fix: kv_bits README docs + address Copilot review on PR #73
github-actions[bot] ccccdeb
docs: expand Supported Models section to full architecture list
github-actions[bot] ed5f8f6
docs: remove GLM 5.1 from supported models (still on feature branch, …
github-actions[bot] File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -1048,9 +1048,20 @@ func handleChatCompletion( | |
| // These are accepted but may not affect generation if MLX doesn't support them | ||
| } | ||
|
|
||
| // ── Validate kv_bits: only nil, 4, and 8 are supported ── | ||
| if let kb = chatReq.kvBits, kb != 4 && kb != 8 { | ||
| let errBody = "{\"error\":{\"message\":\"Invalid kv_bits value \(kb). Supported values are 4 and 8.\",\"type\":\"invalid_request_error\",\"code\":\"invalid_kv_bits\"}}" | ||
| return Response( | ||
| status: .badRequest, | ||
| headers: jsonHeaders(), | ||
| body: .init(byteBuffer: ByteBuffer(string: errBody)) | ||
| ) | ||
| } | ||
|
|
||
| let params = GenerateParameters( | ||
| maxTokens: tokenLimit, | ||
| maxKVSize: config.ctxSize, | ||
| kvBits: chatReq.kvBits, | ||
|
||
| temperature: temperature, | ||
| topP: topP, | ||
| topK: topK, | ||
|
|
@@ -1200,9 +1211,13 @@ func handleChatCompletion( | |
| // raw <|image|>/<|audio|> token embeddings instead of the projected features. | ||
| let isMultimodalRequest = lmInput.image != nil || lmInput.audio != nil | ||
|
|
||
| // Try to restore via token-by-token prefix match (llama-server style) | ||
| // Try to restore via token-by-token prefix match (llama-server style). | ||
| // Skip for quantized-KV requests: the prompt cache stores KV state produced | ||
| // with KVCacheSimple; restoring it into a QuantizedKVCache (or vice-versa) | ||
| // is unsafe and produces incorrect results or runtime failures. | ||
| let skipPromptCache = isMultimodalRequest || params.kvBits != nil | ||
| var stream: AsyncStream<Generation> | ||
| if !isMultimodalRequest, let cachedCount = await promptCache.restore(newTokens: promptTokens, into: cache) { | ||
| if !skipPromptCache, let cachedCount = await promptCache.restore(newTokens: promptTokens, into: cache) { | ||
| // Cache hit: KV state is pre-populated up to cachedCount tokens. | ||
| // Only compute the remaining (new) tokens. | ||
| var startIndex = cachedCount | ||
|
|
@@ -1251,6 +1266,10 @@ func handleChatCompletion( | |
| let onPrefillDone: (() async -> Void)? = { | ||
| if turboHasCompressed { | ||
| print("[SwiftLM] 🧠 Skipping prompt cache save — TurboQuant has compressed \(cache.compactMap { ($0 as? KVCacheSimple)?.compressedOffset }.max() ?? 0) tokens. Saving would decode ~37 GB back to fp16.") | ||
| } else if params.kvBits != nil { | ||
| // kv_bits is set: the cache contains QuantizedKVCache layers whose token | ||
| // format is incompatible with the FP16 KVCacheSimple format expected by | ||
| // promptCache.save. Skip saving to prevent unsafe mixed-format restores. | ||
| } else { | ||
| await promptCache.save(tokens: promptTokens, cache: cache) | ||
| } | ||
|
|
@@ -2305,6 +2324,10 @@ struct ChatCompletionRequest: Decodable { | |
| let chatTemplateKwargs: [String: Bool]? | ||
| /// Top-level thinking override emitted by Aegis-AI gateway | ||
| let enableThinking: Bool? | ||
| /// Number of bits for native MLX quantized KV cache (nil = no quantization). | ||
| /// Only 4 and 8 are supported by the underlying MLX QuantizedKVCache. | ||
| /// Enables `QuantizedKVCache` instead of `KVCacheSimple`. Separate from `--turbo-kv`. | ||
| let kvBits: Int? | ||
|
|
||
| enum CodingKeys: String, CodingKey { | ||
| case model, messages, stream, temperature, tools, stop, seed | ||
|
|
@@ -2319,6 +2342,7 @@ struct ChatCompletionRequest: Decodable { | |
| case responseFormat = "response_format" | ||
| case chatTemplateKwargs = "chat_template_kwargs" | ||
| case enableThinking = "enable_thinking" | ||
| case kvBits = "kv_bits" | ||
| } | ||
| } | ||
|
|
||
|
|
||
Oops, something went wrong.
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
kvBitsis threaded intoGenerateParameters, but the prompt cache currently keys only on token prefix and will happily restore a cached KV state produced with different generation parameters. With a per-requestkv_bits, this can mix quantized and non-quantized cache states across requests, which is unsafe and can lead to incorrect results or runtime failures. Consider includingkvBits(and any other cache-shaping params) in the prompt-cache key, or disabling prompt-cache restore/save wheneverchatReq.kvBitsis non-nil.