llm: dont cap context window limit to training context window (#3988)

This commit is contained in:
Jeffrey Morgan 2024-04-29 10:07:30 -04:00 committed by GitHub
parent 7e432cdfac
commit 7aa08a77ca
No known key found for this signature in database
GPG key ID: B5690EEEBB952194

View file

@ -73,8 +73,7 @@ func LoadModel(model string) (*GGML, error) {
func NewLlamaServer(gpus gpu.GpuInfoList, model string, ggml *GGML, adapters, projectors []string, opts api.Options) (LlamaServer, error) {
var err error
if opts.NumCtx > int(ggml.KV().ContextLength()) {
slog.Warn("requested context length is greater than model max context length", "requested", opts.NumCtx, "model", ggml.KV().ContextLength())
opts.NumCtx = int(ggml.KV().ContextLength())
slog.Warn("requested context length is greater than the model's training context window size", "requested", opts.NumCtx, "training size", ggml.KV().ContextLength())
}
if opts.NumCtx < 4 {